History Std Deviation - How to use for a optimized fuel map?
Moderators: Mangus, robertisaar, dex
History Std Deviation - How to use for a optimized fuel map?
Hi, sorry, i feel like a noob ...
I did some rides and this is an excerpt of the last logs. Std Deviation is displayed and i know and understand mathematical background.
My question is, what are the next steps to be done to generate a new mappig with a better fit to my motor?
The destination map format i am aiming for is different in cell format, see here:
Has anybody a link for next steps what I have to to to get the log input to a optimized map?
Thanks for helping a noob ..
I did some rides and this is an excerpt of the last logs. Std Deviation is displayed and i know and understand mathematical background.
My question is, what are the next steps to be done to generate a new mappig with a better fit to my motor?
The destination map format i am aiming for is different in cell format, see here:
Has anybody a link for next steps what I have to to to get the log input to a optimized map?
Thanks for helping a noob ..
The privilege of a lifetime is being who you are ..
I believe the Buell guys have a spreadsheet to do this somewhere, although I could be mistaken.
My recommendation would be to either alter the hsitory table to match the axis breakdowns in the bin editor (which appears to be ECMSpy). For example, the RPM (x) axis in the history table has different breakpoints than the bin editor. You should make them match, as doing so allows you to match the table cells more precisely.
From there, it might more sense to use either the running average or the hsitory average value for the correction history tables, and apply the correction directly to your fuel map.
xoptimizedrsx, whoc seems to be present on many of the Buell forums, can probably help further.
My recommendation would be to either alter the hsitory table to match the axis breakdowns in the bin editor (which appears to be ECMSpy). For example, the RPM (x) axis in the history table has different breakpoints than the bin editor. You should make them match, as doing so allows you to match the table cells more precisely.
From there, it might more sense to use either the running average or the hsitory average value for the correction history tables, and apply the correction directly to your fuel map.
xoptimizedrsx, whoc seems to be present on many of the Buell forums, can probably help further.
***************************************
TunerPro Author
1989 Trans Am
TunerPro Author
1989 Trans Am
-
- Posts: 5
- Joined: Sat Oct 02, 2010 9:25 am
First switch the view in the drop down tab as follows for your maps(tables)
go to tools , preferences, general tab and place a check in reverse rows and reverse columns.
save that data.
now make sure you have the same rpms in the adx as the bike does. If not edit in the ADX history table to match.
Use History agv.
If you like to use a STD. Dev. A excel sheet to automatically assign the values to a page based from the actual mapping you past into the sheet #1 where as you would past the Dev. values in its corresponding box.
This sheet is Not 100% necessary. you will need to make it. I do have a lot of sheets made to do other things with the values.
We are seeing good results with history avg. once you use the logs please move them(old logs) to a different folder or delete old logs. They will not be needed after ran once and you have created new values to manually input into the map per cell as needed. If you see values that look off and possibly incorrect just Omit these. do another log and use that one for each session. remember one log one correction. dont base it from many logs at one time.
Pace yourself. do small areas at once due to the fact we dont have a huge amount of sensors on the Bikes the data gathers over time.
remember to ride in sections. The more data collected in a single cell the better that log will reflect exactly what the injectors are being told to do based from all sensors active in the adx for that History value.
Last: The adx can be added to and other functions applied to it. I just dont have that much time to figure out more related tables and ect.
Maybe a super guru can add stuff now that all the info is there waiting to be arranged. you could look at some of the car datas and use it as a model to add to the current buell ADX based from buel ecm values. GO slow and test twice. Then verify with others and changes made or added to in the ADX.
I'm always willing to post and host new and modified adx's for people to share.
go to tools , preferences, general tab and place a check in reverse rows and reverse columns.
save that data.
now make sure you have the same rpms in the adx as the bike does. If not edit in the ADX history table to match.
Use History agv.
If you like to use a STD. Dev. A excel sheet to automatically assign the values to a page based from the actual mapping you past into the sheet #1 where as you would past the Dev. values in its corresponding box.
This sheet is Not 100% necessary. you will need to make it. I do have a lot of sheets made to do other things with the values.
We are seeing good results with history avg. once you use the logs please move them(old logs) to a different folder or delete old logs. They will not be needed after ran once and you have created new values to manually input into the map per cell as needed. If you see values that look off and possibly incorrect just Omit these. do another log and use that one for each session. remember one log one correction. dont base it from many logs at one time.
Pace yourself. do small areas at once due to the fact we dont have a huge amount of sensors on the Bikes the data gathers over time.
remember to ride in sections. The more data collected in a single cell the better that log will reflect exactly what the injectors are being told to do based from all sensors active in the adx for that History value.
Last: The adx can be added to and other functions applied to it. I just dont have that much time to figure out more related tables and ect.
Maybe a super guru can add stuff now that all the info is there waiting to be arranged. you could look at some of the car datas and use it as a model to add to the current buell ADX based from buel ecm values. GO slow and test twice. Then verify with others and changes made or added to in the ADX.
I'm always willing to post and host new and modified adx's for people to share.
Mangus & Xoptimizedrsx Thanks for the first advices for me .. I am struggling through and wll, with your help i can enter a next step to my tuned maps! Thanks a lot, I am so happy! It's like Xmas because I can stay tuned and ... smile ... smile ...
Now i have fixed the AFV to 100 % for my testdrives when I do the log. Is this a good idea? I have BUEZDoP0 12-17-2008 and all these maps show the lowered AFV which run into lean and lowest AFV and get stuck here in.
Have you heard about this trouble with german Buells XB from 2008 to 2010? I never heard buell has fixed this problem. They are all much to lean and to hot and have no performance.
Greets from Germany,
Fritz
It will take me some time to dig through this information you gave me and to incorporate and get it sucked, but as I said: I am very happy about response because here in germany the scene sucks ...Xoptimizedrsx wrote:I'm always willing to post and host new and modified adx's for people to share.
I am on this stuff since 2007. And my expierence on the buell is about 25.000 km and now I go into more deep. The first 2 Years i rode the bike without changes.Xoptimizedrsx wrote:Pace yourself. do small areas at once due to the fact we dont have a huge amount of sensors on the Bikes the data gathers over time.
Now i have fixed the AFV to 100 % for my testdrives when I do the log. Is this a good idea? I have BUEZDoP0 12-17-2008 and all these maps show the lowered AFV which run into lean and lowest AFV and get stuck here in.
Have you heard about this trouble with german Buells XB from 2008 to 2010? I never heard buell has fixed this problem. They are all much to lean and to hot and have no performance.
How many history sample counts are required to say that table is worth for the new map?Xoptimizedrsx wrote:The more data collected in a single cell the better that log will reflect exactly what the injectors are being told to do based from all sensors active in the adx for that History value.
I don't understand the necessity to do .. Why? A log is a log and is not changed py any further actions. Or am I wrong?Xoptimizedrsx wrote:once you use the logs please move them(old logs) to a different folder or delete old logs.
I did it and the history tables are reconfigured to the new tables. Now it is perfect! Thank's a lot for answers. Sometimes it saves a tremendous time of erraneous searches and trials to fix.Xoptimizedrsx wrote:now make sure you have the same rpms in the adx as the bike does. If not edit in the ADX history table to match.
Greets from Germany,
Fritz
Last edited by RainMan on Sat Nov 10, 2012 5:37 am, edited 1 time in total.
The privilege of a lifetime is being who you are ..
What is "it?" Reversal of rows/cols only applies to the binary table editors. It will not reverse the rows/cols of history tables.RainMan wrote:Does not work ... It always stays the same ...Xoptimizedrsx wrote:go to tools , preferences, general tab and place a check in reverse rows and reverse columns.
***************************************
TunerPro Author
1989 Trans Am
TunerPro Author
1989 Trans Am
Mangus i got another problem with the adx file. changes in the adcx file do NOT appear in the history tables ...RainMan wrote:Thanks, Mangus, I will have to change manually, with no toggle, doesn't matter.
I can insert columns but i cannnot change the content of them ... no new values appear after save of adxx inside TunerPro !!!
Have a look inside the corresponding file:
http://www.c-work.de/out/BUEZD_V5_4.adx
The tables have been edited directly inside the BUEZD_V5_4.adx, because the changes applied inside TunerPro did not affect any visible changes. I have edited the columns to increasing steps 4250-5000-6000-7000 ... steps inside the adx
But if I reopen Tuner Pro the columns stay confused on old values (where the f...c do they come from?) and trying to re-edit them shows old values of a file which definitely contains new values ...<colval index="0" val="0.000000" />
<colval index="1" val="0.000000" />
<colval index="2" val="400.000000" />
<colval index="3" val="800.000000" />
<colval index="4" val="1000.000000" />
<colval index="5" val="1150.000000" />
<colval index="6" val="1350.000000" />
<colval index="7" val="1500.000000" />
<colval index="8" val="1900.000000" />
<colval index="9" val="2150.000000" />
<colval index="10" val="2400.000000" />
<colval index="11" val="2900.000000" />
<colval index="12" val="3400.000000" />
<colval index="13" val="4000.000000" />
<colval index="14" val="4250.000000" />
<colval index="15" val="4500.000000" />
<colval index="16" val="5000.000000" />
<colval index="17" val="6000.000000" />
<colval index="18" val="7000.000000" />
<colval index="19" val="8000.000000" />
Please help,that's a very confusing behaviour I cannot resolve by my own ...
Is there any old content which stays in a hidden memory? The xdf file is NOT linked to the adx.
Fritz (RainMan)
The privilege of a lifetime is being who you are ..
You have to go into Row axis and Column axis and rename the cells. Take a look at this writeup I did on how to make history tables.
http://www.gearhead-efi.com/Fuel-Inject ... n-TunerPro
http://www.gearhead-efi.com/Fuel-Inject ... n-TunerPro
1990 Chevy Suburban
1994 Buick RoadMaSSter Estate Wagon LT1.
1972 IH 1210 Isky Cammed Balenced 345 TBI
GearHead-EFI.com EFI Conversions and Chip Tuners!
1994 Buick RoadMaSSter Estate Wagon LT1.
1972 IH 1210 Isky Cammed Balenced 345 TBI
GearHead-EFI.com EFI Conversions and Chip Tuners!
Hi EagleMark , now it works fine ... Thanks a lot!
the unmentioned feature when working inside cell or table definitions is a hidden autosort and autoclean. Two rows with 0 0 are not allowed and one value is skipped, the same is if there is a missing entry all unfilled cells get the highest valu entered.
the unmentioned feature when working inside cell or table definitions is a hidden autosort and autoclean. Two rows with 0 0 are not allowed and one value is skipped, the same is if there is a missing entry all unfilled cells get the highest valu entered.
The privilege of a lifetime is being who you are ..