Members Login
Username 
 
Password 
    Remember Me  
Post Info TOPIC: Autotune and Data logging for ECU Editor


Guru

Status: Offline
Posts: 1344
Date:
RE: Autotune and Data logging for ECU Editor


Justin, i have been quiet...but following....too many family projects going on in the last months, thanks for the pics, and tips on the new boxes months ago, i am SUPER interested to beta test for you, got private dyno, and know how to use it....MARC.....A.K.A STOCKER..............!



__________________

09 busa.????? now what....still got what it takes.......!

I got what you need...!
www.poweredbyford.com

www.marc@poweredbyford.com



Guru

Status: Offline
Posts: 1233
Date:

Hey Marc, I was wondering where you had got to! I will be in contact soon when i am ready to go with some stuff to test.

__________________

site_logo_small.png

www.WoolichRacing.comTune your bike to the Limit with our Advanced ECU Flashing Products



Guru

Status: Offline
Posts: 1344
Date:

Ready Justin when you are, the bike has got webs on it......some family issues had to be taken care of first, now i'm ready to get into the bike stuff again, i've converted my stuff to vb 2010, from 2008, and am trying to re-new my reneseas assembler stuff also...the boadrs i shared with you i only got 2 built...but they work fine...I want to start getting involved now i got spare time...LMK...Marc



__________________

09 busa.????? now what....still got what it takes.......!

I got what you need...!
www.poweredbyford.com

www.marc@poweredbyford.com



Guru

Status: Offline
Posts: 1247
Date:

stocker wrote:

Ready Justin when you are, the bike has got webs on it......some family issues had to be taken care of first, now i'm ready to get into the bike stuff again, i've converted my stuff to vb 2010, from 2008, and am trying to re-new my reneseas assembler stuff also...the boadrs i shared with you i only got 2 built...but they work fine...I want to start getting involved now i got spare time...LMK...Marc


Glad to see things getting back in line for you Marc. smile

I have mine torn arpat right now, going to machine the fork tube diameter for McIntosh clamps later this week, then start on a new wiring harness.



__________________


Guru

Status: Offline
Posts: 1344
Date:

Yes John, very messy divorce....ahhhhh...now i got a new place and free time now, been getting my inviroment back up and running, also making some changes also....very glad to be back and involved....smile



__________________

09 busa.????? now what....still got what it takes.......!

I got what you need...!
www.poweredbyford.com

www.marc@poweredbyford.com



Newbie

Status: Offline
Posts: 3
Date:

First off thanks so much for a great tuning tool.

Just started using autotune and noticed my WBO2 gauge doesn't match the ECU Editor screen. It's only off .2-3.
For example if the actual AEM gauge reads 12.9 and the editor will say it's 12.6.

I think I have the voltages set correctly, judging by what the AEM UEGO digital instructions say.
0v = 10.0 AFR
5v = 19.98 AFR (It actually only has 4.99v not 5v in the instructions)

Is this normal?

Thanks
Steve

__________________


Senior Member

Status: Offline
Posts: 196
Date:

4.99v versus 5.00v is only a .2% error. Your output is 99.8% correct. I would not worry about it.



__________________


Guru

Status: Offline
Posts: 1233
Date:

Does the gauge run off an analog output from the wideband or is it digital, if the gauge is analog there can be voltage drops and variances in what it displays.

__________________

site_logo_small.png

www.WoolichRacing.comTune your bike to the Limit with our Advanced ECU Flashing Products



Guru

Status: Offline
Posts: 1247
Date:

See if you can measure teh voltage out from the sensor controller (measure teh putput going to the ECU, and then compare it with the calculated value.

__________________


Newbie

Status: Offline
Posts: 3
Date:

The gauge itself is digital. The 0-5v white wire output from the AEM gauge hooked to the ECU is analog.

Here is the manual if it helps any.. http://www.schnitzracing.com/manuals/AEMWBK.pdf

I'll remeasure the voltages and see if they compare. I did it when I first wired it all up and they matched pretty much. All connections are soldered and ground goes straight to the battery.

I have made 4-5 datalogs/corrections and it has made a big difference in drivability smoothness. And I started with a custom map that I thought was good to start with, But found out it was a little rich.

Again thanks guys for all the time and effort you put into this for all of us to use..

__________________


Guru

Status: Offline
Posts: 1233
Date:

based on the voltage/afr in the manual you could put in 0v = afr 10; 5v = afr 20 into ecueditor and that should be correct.

If you find that the analog output from the wideband is not exactly what you are seeing in the digital gauge you could apply a correction to the above figures. so for example based on the 0.3 AFR offset you mentioned above you could adjust the settings to be 0V = AFR 10.3; 5v = AFR 20.3

Good to hear that you are getting good results with the autotune ;)

__________________

site_logo_small.png

www.WoolichRacing.comTune your bike to the Limit with our Advanced ECU Flashing Products



Member

Status: Offline
Posts: 15
Date:

Autotune and Data logging for ECU Editor works on 600 k8
or it can be only for 1000 and 07 busa?



__________________


Guru

Status: Offline
Posts: 1233
Date:

Yes auto tune will work on k8 600, you can also enable fast baud rate and wideband o2 logging in advanced settings. You may need to check which ms0/ms1 map the auto tune applies to to make sure it is applying to the default for the gixxer.

__________________

site_logo_small.png

www.WoolichRacing.comTune your bike to the Limit with our Advanced ECU Flashing Products



Member

Status: Offline
Posts: 15
Date:

Thank you for your prompt response
well then I'll be waiting for your logo box interface

__________________


Guru

Status: Offline
Posts: 1233
Date:

The log box is still a couple of months away from being available, if you want to get started now you can purchase a USB Interface directly from the Woolich Racing web site http://www.woolichracing.com/USBECUFlashingInterface.aspx

Cheers
-Justin.

__________________

site_logo_small.png

www.WoolichRacing.comTune your bike to the Limit with our Advanced ECU Flashing Products



Veteran Member

Status: Offline
Posts: 27
Date:

...still a couple of month? cry



__________________


Veteran Member

Status: Offline
Posts: 76
Date:

Hey there, I have been doing alot more testing with this, testing both the gsxr and busa bins. It seems like I have run into a few issues that I am presently trying to pinpoint the exact cause of the incompatibility with my setup mainly with the busa bin, so my questions may be caused by the current problems and not related to autotune.

First issue, when the laptop battery dies while logging the bike shuts off, I have found that it starts to hiccup when it reaches critical before shutting off. Maybe we could consider modifying the interface so that 5v can be supplied from another source maybe by relay or something after the laptop shuts off to prevent this? Guess this isnt important for people on the dyno or racers but for road riders I think this can be a pain.

I have also found that when recently using autotune during a 500km round trip my tps/rpm maps seemed to tune up quite well while my iap maps seemed to degrade significantly. Could someone maybe explain the use of the iap deceleration filter as Im unsure as to whether this might help the issue. Thanks

__________________


Senior Member

Status: Offline
Posts: 186
Date:

The IAP deceleration filter simply excludes samples associated to an RPM
value lower than the RPM related to previous sample:

If My.Settings.FilterIAPDecl = True And (logValue.TPS "less then" 11)
And (logValue.RPM "less then" previousLogValue.RPM)
Then
Return False
End If

(I wrote "less then" because of forum filters on characters)

In my personal experience I get better sample quality using the
"transition filters" with a value of 1.0. They exclude samples
when their TPS or IAP values have a 1.0 difference from the previous
value. So basically you remove the first sample that does a jump
from one square cell to another. This reduces the number of samples
but improves how significant they are, it is a trade off between quality
and quantity. Basically a couple of bogus samples can disrupt a cell,
with cascade effect on the surrounding cells due to the smoothing,
I'm trying to avoid this.

The IAP degradation might be due to O2 working in closed loop:
you keep adding fuel while the O2 does its correction regrdrless.
The fix would be to disable closed loop, a workaround is to use
a target IAP that mimic the O2 intention, I posted my attempt
to fix the target IAP previously, with that map I'm not experiencing
degradation, but the values might be different for each setup.










__________________


Veteran Member

Status: Offline
Posts: 76
Date:

Cool thanks looking forward to giving those settings a try, I have a feeling this could possibly be part of the problem. I was under the impression that this closed loop issue was only related to the stock narrowband sensor? Im running a wideband does this still apply?

__________________


Guru

Status: Offline
Posts: 1233
Date:

Yes the closed loop is only an issue when using the stock narrowband sensor, there is no closed loop when running a wideband o2 sensor with the appropriate settings selected in Advanced Settings.

Cheers
-Justin.

__________________

site_logo_small.png

www.WoolichRacing.comTune your bike to the Limit with our Advanced ECU Flashing Products



Member

Status: Offline
Posts: 6
Date:

Sorry if I am a bit late in here and probably have not read everything.

I have recently started using Autotune with my 1441cc B-King.
This is of course not with stock exhaust. Brocks header + Woolich linkpipe + Yoshi endcans.
*And* with a wideband O2 sensor + Innovate LC1 (the MTX-L failed after to sessions).

I cannot see the rationale in trying to autotune with stock (narrowband sensor) and closed loop.
I have tried this with PCV+Autotune on my ex Bandit 1250.
*IT will NOT WORK*.

The whole purpose of Autotune is pointless if the ECU at the same time is altering fuel mapping according to lambda=1 (closed loop).

With the Big Bandit which unfortunately is a Suzuki model with an "unhacked" ECU the only way to to Autotune is to do this in open loop/TPS map (where it works OK).

For any other bike/"hacked" ECU it should work OK though as long as Open loop is disabled using a Wideband sensor.

Still, as Ballad has pointed out, you will have to set up filtering as "close ratio" as possible as there obviously is logic in the ECU to alter fuel when deceleration is deteced (sudden drop of IAP).

I have now done a few more sessions applying Autotune corrections and engine response/cruise "smothness" has improved. I also can observe the changes on my Innovate AFR-gauge after applying an Autotune.

My manual corrections (due to engine tuning) in TPS map was farily spot on according to the logs I've done so far.

Autotune is a great feature!!smile




-- Edited by B-King_2009 on Sunday 6th of May 2012 07:32:05 PM



-- Edited by B-King_2009 on Sunday 6th of May 2012 07:34:29 PM

__________________


Member

Status: Offline
Posts: 9
Date:

Hi all.

Have anybody noticed when loading a logged data file the values seem to show up in wrong column? I have been tweaking over and over again with the IAP map using autotune. It starts quite well, but gets worse as the iterations take place. I have done some tests lately trying to find out whats happening rather than just start all over again from scratch. Have been playing around with the Exhaust Gas Velocity Offset to see if there was something there, made a .bin with some very rich spots and they clearly show up one column to the right. Have gone through many previous .bin with their corresponding log files and the same can be observed. E.g. the result of a rich .bin coulomb 23 will be found in Engine Data Viewer coulomb 20 when its data file has been loaded.

I have done all possibly user errors over the last couple of years and learn a lot using EE and following the board. It`s great and it`s fun. For the record, I have a wideband and running EE 2.5.5.110. Question is, what am I doing wrong or is it a bug causing this?



__________________


Senior Member

Status: Offline
Posts: 186
Date:

rolnos wrote:

Hi all.

Have anybody noticed when loading a logged data file the values seem to show up in wrong column? I have been tweaking over and over again with the IAP map using autotune. It starts quite well, but gets worse as the iterations take place. I have done some tests lately trying to find out whats happening rather than just start all over again from scratch. Have been playing around with the Exhaust Gas Velocity Offset to see if there was something there, made a .bin with some very rich spots and they clearly show up one column to the right. Have gone through many previous .bin with their corresponding log files and the same can be observed. E.g. the result of a rich .bin coulomb 23 will be found in Engine Data Viewer coulomb 20 when its data file has been loaded.

I have done all possibly user errors over the last couple of years and learn a lot using EE and following the board. It`s great and it`s fun. For the record, I have a wideband and running EE 2.5.5.110. Question is, what am I doing wrong or is it a bug causing this?


Which bike are you using (gen2, bking, gsx-r)? The target AFR wich comes with EE is not ideal for IAP (IMHO). To build my target IAP map

I start from stock bin, do a long session and then set the target values equal to the logged values if larger then 13.2 in the first 6 columns,

with minimal smoothing (so values are always decreasing whem rpm increases).Also PAIR must be closed, ignition and stp maps unified

otherwise gear changes cause nothing but data mixing in the logged values. In this way %map change should be close to zero

in IAP top left corner for all following iterations (this applies especially for stock o2 sensor).

Here my optimal target IAP afr up to 10000 rpm (bking stock everything):

14.614.214.013.813.613.413.213.213.213.213.213.213.213.213.213.213.213.213.213.213.2
14.614.214.013.813.613.413.213.213.213.213.213.213.213.213.213.213.213.213.213.213.2
14.614.214.013.813.613.413.213.213.213.213.213.213.213.213.213.213.213.213.213.213.2
14.614.214.013.813.613.413.213.213.213.213.213.213.213.213.213.213.213.213.213.213.2
14.614.214.013.813.613.413.213.213.213.213.213.213.213.213.213.213.213.213.213.213.2
14.614.214.013.813.613.413.213.213.213.213.213.213.213.213.213.213.213.213.213.213.2
14.614.214.013.813.613.413.213.213.213.213.213.213.213.213.213.213.213.213.213.213.2
14.614.214.013.813.413.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.2
14.614.214.013.813.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.2
14.614.214.013.813.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.2
14.614.214.013.813.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.2
14.614.214.013.413.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.2
14.614.214.013.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.2
14.614.214.013.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.2
14.614.214.013.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.2
14.614.214.013.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.2
14.614.213.813.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.2
14.614.213.813.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.2
14.614.013.613.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.2
14.614.013.613.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.2
14.614.013.413.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.2
14.414.013.413.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.2
14.414.013.413.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.2
14.414.013.413.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.2
14.214.013.413.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.2
14.214.013.413.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.2
14.214.013.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.2
14.214.013.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.2
14.214.013.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.2
14.214.013.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.2
14.014.013.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.2
14.014.013.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.2
14.014.013.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.2
14.014.013.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.2
14.014.013.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.213.2



__________________


Member

Status: Offline
Posts: 19
Date:

This is exactly what I want to do on my K11 Stage 1 (303 hp). I have the new datalogger and I flashed my previous non-boosted 'busa. So it's not new, I just want to make sure I don't hurt this bike considering the additional variables boost brings. My K11 Stage 1 has a bit of a rough idle and when coming off throttle can spit and carry on.

Can you share you procedure on this? I can get all my maps loaded withouth trouble I'm just want to make sure I only flash the non-boosted section, though.

In the interest of full disclosure, my build was done my Smithers in KC, and he wasn't comfortable going fully with the ecu, so there is a PC5 also.

jgennaro wrote:


Justin,

Now that I got the smoothing thing figured out I've made 8 or 9 runs today and the boost filter worked perfect. In addition, being able to filter the boost made tuning the TPS map much easier.

The bike off boost is much more crip and responsive. Thank you for all of your help again.

Jason




__________________

2K9 Black 'Busa - DOA

K11 Black 'Busa RCC Stage 1 - 303 HP

2009 1989 ZX9R - labor of love



Member

Status: Offline
Posts: 9
Date:

Thanks for the reply Ballad2, interesting approach you describe making your IAP map.

I have a Gen2 all stock exept full Yosh R77(PAIR is off).AHave made many .tafr`s and flashed 100 times in my attempt to get everything as I want.Good respons on low rpm, smooth engagement of torque when on gas, fuel economyand to a cirtian extend clean combustion.The stupid thing is that I havean AFR gauge and a TPS dial(useful before the Logbox) so I keep watching these while driving, and then I figure out that things could always be better, even I am quite happy as it is.

What I don`t understand is why the resault of a certian IAP map cell e.g. 3000/27will be found in next right cell 3000/25 when looking at the log in Engine Data Viewer. Consequently the map can never match the .tafr when autotune.



__________________


Senior Member

Status: Offline
Posts: 186
Date:

The IAP target values for all column 27 and 25 should be equal (i.e. 13.0 or 13.2 all rpm). In my last busa logs, total

of about half million samples, IAP column 27 and 25 averages are very close to each other and to the target as well.

Maybe you just notice a difference between istantaneous and average values.

Try to redo the autotune filtering using only values in the 12.0 to 15.0 range and set IAP decel and transision filters.



__________________


Member

Status: Offline
Posts: 9
Date:

I don`t think you understand. There is nothing wrong with my .tafr or map (for academically reasons I just want to have it "spot on").

Let me explain in another way. I load my .bin into EE and manually edit the IAP map (which is fine and close to my.tafr)just for a test purpose.I make, lets saycolumn 27 really fat by adding 50%, all RPMs. I flash thistest .bin and go for a long ride and log. Later when loading this .wrl logfile into Engine Data Viewer I can clearly see the resault of this very rich IAP map cells, a nice deep blue column of afr figures in the 10-11 range, but they are not in column 27. They are all in column 25! If I now Autotune this(correct .tafr is loaded) the program will decrease fuelfigures in all this cells in column 25 which is understandable, but as we will see it`s all wrong.OK, I autotune and save the file. Open the file in with "Edit Fuel Maps" in EEwith my original .bin as Compare map. Columb 27 is still 50% more fat than the compare map, in addition column 25 is obviously a lot leaner.

I have tryed to alter the Exhaust Velocity Offset to see if I by this can "place" the rich testfigures in the right place, but no. Have anybody tested what I have explained? Until I can sort this out I must tweak my map manually. Autotuning just make myit worse.



__________________


Senior Member

Status: Offline
Posts: 186
Date:

rolnos wrote:

I don`t think you understand. There is nothing wrong with my .tafr or map (for academically reasons I just want to have it "spot on").

Let me explain in another way. I load my .bin into EE and manually edit the IAP map (which is fine and close to my.tafr)just for a test purpose.I make, lets saycolumn 27 really fat by adding 50%, all RPMs. I flash thistest .bin and go for a long ride and log. Later when loading this .wrl logfile into Engine Data Viewer I can clearly see the resault of this very rich IAP map cells, a nice deep blue column of afr figures in the 10-11 range, but they are not in column 27. They are all in column 25! If I now Autotune this(correct .tafr is loaded) the program will decrease fuelfigures in all this cells in column 25 which is understandable, but as we will see it`s all wrong.OK, I autotune and save the file. Open the file in with "Edit Fuel Maps" in EEwith my original .bin as Compare map. Columb 27 is still 50% more fat than the compare map, in addition column 25 is obviously a lot leaner.

I have tryed to alter the Exhaust Velocity Offset to see if I by this can "place" the rich testfigures in the right place, but no. Have anybody tested what I have explained? Until I can sort this out I must tweak my map manually. Autotuning just make myit worse.


I understood the kind of experiment you did, I was trying to interpret it. First of all it would be useful to understand if this behavior is

limited to a specific column shift (that could imply a programming bug) or related to a range of columns.

A consistent shift among different columns could be caused by the IAP sensor calibration/fluctuation/granularity.

The map value is a base, but the ECM applies several other level of corrections before calculating the effective fuel pulse.

These corrections are designed to work well for the average case and stock map, but could produce unexpected results when probed

under untested artificially created conditions.

For a better analysis of your experiment, using your existing data you could try to set the iap transition to 1.0,

so the drift between colums is removed and we would be sure to observe only sequences of samples in the same column.

The exhaust delay play a role only during cell transitions, I'd like to see the result when transitions are filtered ahed of any time delay setting.

That said the smoothing algorithm is designed to propagate the effect of samples to surrounding cells, assuming the regularity of the

3D AFR surface as a function of (IAP, RPM) and the trust in the sample values. In your case setting artificially wrong values has anegative

propagaiton effect on the adjacent columns.

In this experiment only logged vaules (and %map change) are relevant, not the map resulting from autotune.



__________________


Guru

Status: Offline
Posts: 1233
Date:

Hi Rolf,

I just tried checking to see if values are being places in the wrong columns and it appears to be ok to me? can you send me your bin file and the log files and some pics explaining which cells i can look at to better understand your problem?

Cheers

-Justin.



__________________

site_logo_small.png

www.WoolichRacing.comTune your bike to the Limit with our Advanced ECU Flashing Products



Member

Status: Offline
Posts: 19
Date:

I have the new log box on a k11 Hayabusa with an RCC Stage 1. It has water injection and an ams controller. My tunere (Steve Smithers) wasn't confomfortable flashing, so he also hung a PCV on my build.

I had flashed my K9 NA bike many ties with no issues. What I'm hoping to do here is to flash the rpms at idle, and off boost. I understand there is a way to do this, but I've not yet been able to put it together.

The bike spits and spudders when at idle, this is what I'd like to fix formost. Any suggestions or how to flash without blowing stuff up?

__________________

2K9 Black 'Busa - DOA

K11 Black 'Busa RCC Stage 1 - 303 HP

2009 1989 ZX9R - labor of love



Guru

Status: Offline
Posts: 953
Date:

why does it spit and sputter at idle? I assume you are running stock fuel pressure? Should idle very close to stock unless you have some really wild setup or something plumbed to IAP sensor incorrectly etc.

__________________


Member

Status: Offline
Posts: 19
Date:

It'll take 2 atempts to start it, and then setting still the rpms change 300 or so. It also like to die when comming off throttle for like a light that just turned red.

The maps I've loaded show quite a bit of changes they would make, I just want to make sure I'm not affecting the boost map before I proceed.

Fuel pressure is 32-35.



__________________

2K9 Black 'Busa - DOA

K11 Black 'Busa RCC Stage 1 - 303 HP

2009 1989 ZX9R - labor of love



Guru

Status: Offline
Posts: 953
Date:

Why is fuel pressure so low?

__________________


Member

Status: Offline
Posts: 19
Date:

I dind't realize 35 was. Idling it my warry between 32-35, but when rolling it is usally 36. What should it be?



__________________

2K9 Black 'Busa - DOA

K11 Black 'Busa RCC Stage 1 - 303 HP

2009 1989 ZX9R - labor of love



Guru

Status: Offline
Posts: 953
Date:

stock is more like 42-43, which is ~30% more pressure then 32 psi, and about ~14-15% more fuel flow than what you have (assuming your gauge is accurate). Is this a fmu setup?



-- Edited by smithabusa on Thursday 16th of August 2012 03:22:01 PM

__________________


Member

Status: Offline
Posts: 19
Date:

Greg, the FMU question always messes me up. How would I tell?



__________________

2K9 Black 'Busa - DOA

K11 Black 'Busa RCC Stage 1 - 303 HP

2009 1989 ZX9R - labor of love



Member

Status: Offline
Posts: 9
Date:

Hi Justin,

Sure, I`ll pack some files and send you.

Thanks,

Rolf



__________________


Member

Status: Offline
Posts: 19
Date:

I believe I'm using the riding rate FMU that can with the RCC kit, but I'm not sure. I have a call into Steve.

Is it possible the reading is coming before the FMU? when running, it's allways pretty solid at 36



-- Edited by Craash on Thursday 16th of August 2012 04:50:38 PM

__________________

2K9 Black 'Busa - DOA

K11 Black 'Busa RCC Stage 1 - 303 HP

2009 1989 ZX9R - labor of love



Member

Status: Offline
Posts: 6
Date:

rolnos wrote:

I don`t think you understand. There is nothing wrong with my .tafr or map (for academically reasons I just want to have it "spot on").

Let me explain in another way. I load my .bin into EE and manually edit the IAP map (which is fine and close to my.tafr)just for a test purpose.I make, lets saycolumn 27 really fat by adding 50%, all RPMs. I flash thistest .bin and go for a long ride and log. Later when loading this .wrl logfile into Engine Data Viewer I can clearly see the resault of this very rich IAP map cells, a nice deep blue column of afr figures in the 10-11 range, but they are not in column 27. They are all in column 25! If I now Autotune this(correct .tafr is loaded) the program will decrease fuelfigures in all this cells in column 25 which is understandable, but as we will see it`s all wrong.OK, I autotune and save the file. Open the file in with "Edit Fuel Maps" in EEwith my original .bin as Compare map. Columb 27 is still 50% more fat than the compare map, in addition column 25 is obviously a lot leaner.

I have tryed to alter the Exhaust Velocity Offset to see if I by this can "place" the rich testfigures in the right place, but no. Have anybody tested what I have explained? Until I can sort this out I must tweak my map manually. Autotuning just make myit worse.


Did you get a reply from Justin?

What you describe here is pretty much spot on with my (first) tries with Autotune+Wideband sensor+LogBox.

After turning off most filters/autotune(smoothing) and just analyze the logs (Logdata viewer) I reach my target AFR in 3-4 sessions with manual changes according to suggested percent "off" target AFR from Logdata viewer (from a base bin/engine mods).

Fiddling with Exhaust Gas offset makes things worse like you describe.

I tried to enter (correct) values for my Brocks headers and this makes the logged data "offset" from actual cells(IAP map).
Setting offset time (ms) to 0 works best for me.

@Ballad
A target AFR map is nothing more than whatever you want your AFR to be.
It may be "wrong" regarding what is best for engine response/longevity/optimum AFR for whatever running condition.
If you want most mgp (economy) in IAP/cruise range you try to set/get the AFR for this.
To set target AFR to observed AFR is not the way to "Autotune".
Neither is use of stock logic/stock sensor for closed loop (also known as "fuel trim") regarding use of EE Autotune functionality.
If the ECU/injectors(a working sensor) can correct for observed AFR to target AFR (closed loop=14.7) you can set "any" value for injector time in closed loop map.
(I hope you see the point that the ECU will maintain 14.7 in closed loop regardless with what you set in target AFR, unless something wrong like a vacum leak is present).
The ECU will subtract/add from/to this "base" to achieve "closed loop=14.7 or the ECU will "throw" a DTC (eg 0171 if the ECU was OBD compliant).
This is pretty basic and has been like this for decades in the automotive industry (different ECU's/"firmware).
In the automotive industry the ECU's will also alter "target AFR" related to working condition ("boost" condition in a 1.8T VAG as an example).
All aftermarket Autotune-"boxes" works the same way(regarding adding/subtracting from a base map) regarding fuel injected motorcycle engines/ECU "piggy-backs".

If some of the members here programmes ECU's for the automotive/motorcycle industry it would be nice if they could shed some inside information in this discussion. :)



__________________


Member

Status: Offline
Posts: 9
Date:

Hi B-King 2009.
Justin couldn`t look into the material I send him right away, no time due to variouse reasons. I actually told him to leave it since the issue I described is no problem for me. I know what to do and obviously, so do you. Like you saied I also reached my target AFR after 3-4 iterations by manually editing the IAP map. Never been so close to target before using autotune, and I have used it many many times since Justin made it available more than two years ago.

I have made lots of target AFRs in my attemts to get best driveability in respect of performance, response, comfort and MPG. I am not racing so it`s all for academically reasons, and it is fun to use this great tool changing things to the better. I drive a lot, average 15K km/year (just came back from a 6000 km / 8 days tour). 95% of this is on IAP map, so thats why this is most important for me to have as I want. My target IAP AFRs (latest and gratest) changes across the table, from 13 to 14,6 RPM- dependent and from 12 to 14,6 vacum-dependent. The TPS target take over from 14,6 at 11% falling to 13,6 at 30% - 100% from 4000 and up. More rich under 4K.

The challange have proven to be that "things change" for no reason. I guess the accuracy and repetability of the sensors are not that good. I have a LC1 wideband and for sure it is not the best you can get. Different brands and blends of fuel effects the AFR also. The APS compensating seems to work great, can not see significant changes watching the AFR gauge while climbing high mountains (did Col de la Bonette a few days ago, highest pass in Europe, France -2860m). What really makes a big differece though is the temerature increase in the throtlebody due to radiated heat from the engine. This will not be compensated since the IAT sensor sits upstream in the airbox. To compensate this I use MS map (stock - much more rich than my own) the first 10 - 15 minutes, works just fine.

Regards,
Rolf





__________________


Senior Member

Status: Offline
Posts: 149
Date:

Justin

Just finished my mates 2010 busa with Trim 57 Turbo.
Data logging does not give me an option to Autotune the Boost fuel map.
Also the pressure scale differs from the Log to Boostfuel map.

It sure would save a hell of a lot of time if we could autotune Boost Fueling on the Dyno for starters and then on the road/track for fine tune using Logbox.
Or is there a specific reason for not having this functionality (risk ?)?

BTW using Zeitronic Wideband.


__________________


Guru

Status: Offline
Posts: 1233
Date:

Hi Len,

you are right, there is currently no option to auto tune the boost fuel maps. I have had a few different options on the best way to do this, and i am open to some suggestions. I will be implementing the data logging and auto tune in the Woolich Racing Tuned software soon so its a great time to start a discussion about this. What are your thoughts on how it should work?

Cheers

-Justin.



__________________

site_logo_small.png

www.WoolichRacing.comTune your bike to the Limit with our Advanced ECU Flashing Products



Guru

Status: Offline
Posts: 1247
Date:

Tuing teh boosted areas of the fueling is a tricky one.

Theoretically, the boosted areas should require fuel proportional to the density ratio of the inlet air charge. This holds fairly true, if you have base map, in this case an Alpha-N (TPS vs RPM) map.

The issue is that the base motor map requirement changes quite a bit when you start lowering copression and adding all of the restriction of the inlet plumbing and the turbo housing. I have been thinking about how to account for this, especally since my application is very different (completely different ECU and only 1 set of injectors).

Currently the "standard" way to tune is to leave the base maps alone and modify the fueling in theboosted areas. That said, Greg Watters has noted that he has neem using a theoretical boost fueling map and then tunes with the TPS.

I have tried testing without boost, but did not have success as I hav no charge tube connected at all and there was just too much restriction. My future testing will hopefylly be with a removed wastegate adn connected charge tube to get a better "base map". Presumably the boost fueling will then have a more consistent shape, more closely following the theoretical density ratio changes.

I also plan to add an intercooler to my setup, which will throw another change into using the intake pressure alone to try to determine the fueling (350F air at 35psi has much less density than 200F at 35psi). The best way is likely to add in a realistic intake temperature curve as well.

This didn't really directly add much to the topic of auto tuning, just wanted to throw out some other thoughts in general.



__________________


Guru

Status: Offline
Posts: 1233
Date:

good discussion John, this sort of stuff is really interesting. you mentioned to me previously about how other systems have a simpler/different approach to boost fueling would this simplify the auto tune for boost fueled bike. Greg Smith had some good ideas around boost fueling and auto tune as well, i will ask if he can contribute his ideas here ;)



__________________

site_logo_small.png

www.WoolichRacing.comTune your bike to the Limit with our Advanced ECU Flashing Products



Guru

Status: Offline
Posts: 546
Date:

Justin the path already begun with editor probably just needs refining, as making a total change in the system requiring a complete retune may stop some from changing to your software

what is there now works reasonably well, , not sure my theretical table thing worked as well as i hoped , it made tuning easy with autotune in tps , but not interchangable between bikes very much as they all differ in effiancys

Thinks i would like to see , ign retard in actual degrees , no extra popup boxes, like the current promps when changing com port, or opting legacy/new boost by selecting ok or cancel there realy annoying

Can't it be made to autoselect ?

trouble with autotune is being absolutely sure the sensor calibration is correct, and that your isolating misfires ,



__________________


Senior Member

Status: Offline
Posts: 149
Date:

Just got back from first Speedweek in Africa.
This is the place they plan to run Bloodhound Landspeed Record next year.(Speedweeksa.com)
They are planning a 1000mph run , and they think we are crazy ?
It was a bit disapointing and the cars tore up the track completely by the time we got there.
Bikes were very unstable , drifting all over the place beyond 250km/h , wind gusts were also doing thier best to kill us , decided to rather live to fight another day.
Doing high speeds on clay track sure has a steep learning curve.
Apparently they will prepare better next year when the Bloodhound Poject will run also.

Maybe it will attract some Crazy Australians ??

Will give some feedback / thoughts regarding Autotune for Turbo in the week as we tune the current bike.

__________________


Guru

Status: Offline
Posts: 546
Date:

Len welcome to going fast on a loose surface ;)

the 250kph range is a bit unstable, if you can blast beyond it its not bad, but if your stuck there its unnerving and hard to be confident it will improve.
Wind is difficult to deal with, you can't lean into it very far, 5-6 mph is realy the limit , or it just pushes you across the track
Cars will tend to use the middle of the track so hug the sides
Carry weight, i had nearkly 300lbs on the busa , swingarm full of shot, weight box between the tire and swingarm (+8 over swingarm)
and more bolted to the sides

And we are discussing the logistics of getting the bikes there , major questions revolve around getting the bikes and people in and out safely,

need to find out more about your ports and customs services



-- Edited by Maj750 on Monday 1st of October 2012 01:05:44 AM

__________________


Senior Member

Status: Offline
Posts: 149
Date:

Greg

I sh*t you not (12mph crosswinds), it scared the beejesus out of us all.

Justin ,

I have been giving the autotune for Boost some thought.

Tuning the boostfuel in the boost fuel map will be better than in the TPS map ,
considering boost per gear , different loads on motor in gears etc. finetuning TPS should be OK.

Logging is fine as it is , just the pressure scale should correlate with Boost Fuel Map pressure.
That is to say it should maybe be adjustable for 3 bar map sensor , psi/kpa etc. (same adjustments as boost fuel)

Once this lines up I cant see why it coud not function as normal TPS autotuning ?

When tuning one can start with small boost increments and conservative AFR to get it in the ball park.
From there Datalog on road for finetune and TPS map adjustments.

What do the other more experienced guys think ?



-- Edited by LenTurbo on Monday 1st of October 2012 10:25:10 AM

__________________


Guru

Status: Offline
Posts: 546
Date:

LOL there the ones that blow you across the track while under power , but if you roll off slightly you can regain some position on the up wind side

Boost fuel autotune should be able to work but i wonder about failsafes , like a change in user level to access the function ??


__________________


Senior Member

Status: Offline
Posts: 149
Date:

Finished of current turbo bike tuning last night.
This is what we got with 95 Pump gas @ 10psi boost - 309 rwhp and 220nM !!
This is with very conservative settings as follows :

Timing reduced to 26 Deg. across board from 6000 Rpm and 50% throttle all the way to 12k Rpm and 100% throttle.(This requires more attention)
Autotuned IAP map on dyno.
Autotuned TPS map up to 40% throttle on dyno.
Ran with logbox to 3psi boost and made manual adjustments in Boostfuel Map (Multiple runs).
Ran with logbox to 7psi boost and made manual adjustments in Boostfuel Map (Multiple runs).
Ran with logbox to 10psi boost and made manual adjustments in Boostfuel Map (Multiple runs).
Under varying boost the bike sits consistantly at AFR 12:1 - nice and rich for dyno runs.
This whole process took about 2 hours.

With autotune function in Boost fuel this time will be cut in half.
Will run datalogging on road this week to fine tune.
Any suggestions for timing map ?
We put in a 2mm base gasket , compression should be at +- 9:1
Bike has no intercooler.




__________________
« First  <  Page 18  sorted by
 
Quick Reply

Please log in to post quick replies.

Tweet this page Post to Digg Post to Del.icio.us


Create your own FREE Forum
Report Abuse
Powered by ActiveBoard