Hey everyone,

I have a few magical spreadsheets that I've put together to make some aspects of tuning a bit easier to visualize and calculate.

I'm looking for around 3 people, preferably those using the ECUFlash Standalone Logger, to volunteer to try a few things out. Any stage in your tuning is fine, I just want to make sure that it is easy enough to use without having me standing over your shoulder. And finding issues, obviously.

EDIT: If you're looking for the spreadsheet(s), use the most recent one. Surf the thread backwards if you must.

I volunteer!

count me in mate . . . u have my emai still rite ?

Not using the standalone logger but I would give it a shot anyways.

I've converted a map smoother to work with the X/RA Map size, it's not perfect yet but working on it.

Great! Thanks guys.

I'll post up what I have ready in this thread, give it a go if you have a chance. Anyone can test it, but if a few of you really take a look and deliberately look for errors, I'd appreciate it.

kasander, I am interested in seeing what you're referring to!

PM me your email, I'll send you the file.

It's a bit messy tho but you'll understand quickly :)

Ok, initial test run. I've attached my Standalone Logging Analysis and MAF Scaling Template.

Try them out as you wish, let me know if you run into issues or have any questions. I haven't done all of the writeup on "instructions", so hopefully it's not too poorly designed.

I've spent enough time using these that it all makes sense to me, which is why I need outside opinions. Take the time to add your tables from ECUFlash. It's tedious but worth it (I hope).

PM me your email, I'll send you the file.

It's a bit messy tho but you'll understand quickly :)

ForumName atgmail dotcom

A "To Do" so far has been to properly implement LTFT Idle and Cruise and their switchover points into the MAF Scaling Template.

I had followed Hollywood's description, which basically just uses the sum of STFT and LTFT (Cruise), but in actual fact the idle should be used at lower MAF Voltages as well. I'll correct soon, but really it doesn't make too much of a difference as of now.

I took a look at the MAF scaling template and I noticed on the first tab you say to log RPM, LTFT_Idle, LTFT_Cruise and STFT. but on the next tab you have MAF_Volts instead of RPM..... im guessing MAF_VOlt is what we want?

and also, for the values in the MAF tables, i know mine show up as 419.12 not 41912.... is there a difference? or are the decimals missing? different roms?

I took a look at the MAF scaling template and I noticed on the first tab you say to log RPM, LTFT_Idle, LTFT_Cruise and STFT. but on the next tab you have MAF_Volts instead of RPM..... im guessing MAF_VOlt is what we want?

and also, for the values in the MAF tables, i know mine show up as 419.12 not 41912.... is there a difference? or are the decimals missing? different roms?

Thanks Storm,

Yes, Volts is what we want. I'll adjust the instructions for the next one. Thanks for the heads up!

The MAF Tables I used are the ones specifically indicated in the document. If you're using a different address/format, it will show up differently. I matched the ATR tables formatting so it would work for both. If you want to use your own scaling tables, just divide by 100.

Thanks Storm,

Yes, Volts is what we want. I'll adjust the instructions for the next one. Thanks for the heads up!

The MAF Tables I used are the ones specifically indicated in the document. If you're using a different address/format, it will show up differently. I matched the ATR tables formatting so it would work for both. If you want to use your own scaling tables, just divide by 100.

awesome! I will deff give it a try and report back in! I am in need of some MAF scaling help!

So I guess to work with your spread sheet, I would copy my values, multiply by 100, past them into the sheet with my log. Get new values, and then divide my 100 and past back into my rom?

Im up for some R&D :D

EDIT

Whats your definition of long? like 20 minutes? or an hour?

wow! appreciate the effort you put into this! :thumbup: i'll have to take a deeper look when i get home and have some logs handy.

awesome! I will deff give it a try and report back in! I am in need of some MAF scaling help!

So I guess to work with your spread sheet, I would copy my values, multiply by 100, past them into the sheet with my log. Get new values, and then divide my 100 and past back into my rom?

Im up for some R&D :D

EDIT

Whats your definition of long? like 20 minutes? or an hour?

I'm usually about 20 minutes. For the MAF, you should technically disconnect the battery to reset fuel trims, THEN go for your drive. Another solution would be to use my definitions ;)

wow! appreciate the effort you put into this! :thumbup: i'll have to take a deeper look when i get home and have some logs handy.

:thumbup:

I'm usually about 20 minutes. For the MAF, you should technically disconnect the battery to reset fuel trims, THEN go for your drive. Another solution would be to use my definitions ;)



:thumbup:


hmm maybe i should change then? I am using what Clipse posted in his "how to" for MAF Scaling. I have the same 52680015.xml inherit in my car.

Im confused tho, how could the same XML's have different definitions? Multiple ways to skin a cat i guess?

Yeah, I posted one of the sets in Clipse's HowTo ;)

I just changed it to a more "workable" version more recently.

And you're right on, they are just multiple ways of doing the same thing. I have used both to do the same thing, with success in each. Try switching your MAF Scaling stuff with mine for the time being, if only long enough to test the spreadsheet :)

Yeah, I posted one of the sets in Clipse's HowTo ;)

I just changed it to a more "workable" version more recently.

And you're right on, they are just multiple ways of doing the same thing. I have used both to do the same thing, with success in each. Try switching your MAF Scaling stuff with mine for the time being, if only long enough to test the spreadsheet :)


Will do :thumbup:

Thanks for the hard work!

I will post back hopefully this weekend as I am at work right now!

Yeah, give it a try. I want to change something in it, I'd like to use the "current' LTFT instead of Cruise, so that it is calculated precisely.

If anybody has a 0020 ROM, can you tell me the address for MUT50? I don't have my computer.

It's 8088A9, what's that for?

It's nice your MAF scaling table but I don't have the adress to log MAF Volts on the GSR 2010 (55570006). Does anybody got it or the MUT reference for it?

Thanks!

Do you really have 55570006? I have 55570005. I wonder what was changed.


Edit: I just noticed you are in Canada. That could be the diff.

It's 8088A9, what's that for?

It's nice your MAF scaling table but I don't have the adress to log MAF Volts on the GSR 2010 (55570006). Does anybody got it or the MUT reference for it?

Thanks!

Thanks, I'll try it when I get back in town.

It (should) be the "Current" Fuel Trim, which should be a more accurate method of calculation, rather than relying on Cruise alone.

I'll check what I have for your Volts when I get home as well. Post up if you find it in the meantime though, it would be helpful. You could open an older ROM, backtrack through the MUT table and figure it out on the new ROM.

FWIW: Cobb seems to use something like Current Fuel Trim, because all you can log is LTFT. While driving, when you let off the gas, you can see the trims hard switch to the Idle Fuel Trims, then as soon as you step on the gas again, it will jump right back to Cruise Fuel Trims.

Yeah, I think I like that idea better. As soon as I can test it, I'll update.

Do you really have 55570006? I have 55570005. I wonder what was changed.


Edit: I just noticed you are in Canada. That could be the diff.

Don't know what changed but if you want to upgrade to that rom, I can send you my original untouched file via email.

I think Bryan said even 006 was not the latest but it would not hurth to play with a newer file.

I'll check what I have for your Volts when I get home as well. Post up if you find it in the meantime though, it would be helpful. You could open an older ROM, backtrack through the MUT table and figure it out on the new ROM.

I've tried searching in a 0020 file but I can't find the adress entered in EvoScan. It's probably an offset so no luck so far.

You can be sure I'll post it if I find it, no reason not to share, it's only a logging adress :rock:

Thanks, but right now I will stick with what I have.

MUT50 is the current LTFT in use..

MUT50 is the current LTFT in use..

Yes... That's what I meant?

I just wanted the address, since I don't have my PC here. I'll be home soon!

Ok, MAF spreadsheet is good to go! Now uses LTFT Current and has my configuration for the standalone logger in case you just want to use that.

I've only tested it on a 2008 GSR, but I'm 94% sure it will work on an MR. Golden and other 2010 people may need to find new addresses and maybe even definitions for scaling. Sorry. If you ever do find it, let me know and I'll add the information.

I haven't heard anything from anyone, so I still don't know if these are useful, helpful or garbage...

What Addresses do you want to know? I have a handful.

For Logging, the easiest way to get your addresses is to use the MUT table to look up the addresses.

MUT0C = LTFT Low (aka Idle)
MUT0D = LTFT Mid (aka Cruise)
MUT0E = LTFT High (Not really used because of the way it's set up in a table Tephra posted in the "Disable LTFT trims from affecting Open Loop" thread)
MUT0F = STFT
MUT50 = LTFT_In_Use - the currently used Fuel Trim. This is what Cobb logs when you select LTFT. It will swap between LTFT Low and LTFT Mid when you go from idling to cruising (and vis-versa).


I have found some stuff for the 55570005 ROM. (MAF Scaling and stuff)

Thanks Golden.

Download the spreadsheet, I have all the MUT information in it already in the "Information" tab. The things I don't have is the MAF Scaling information for 2010 people. I'm sure it would benefit people wanting to scale!

Oh, I see. I'll take a look and report back.

Oh, I see. I'll take a look and report back.

Thanks! The more eyes the better. :thumbup:

Here is what you will need if you want more in there:
2010 Evo X GSR ROM 55570005
And
2010 Ralliart ROM 55590006


<table name="MAF Scaling Part 1" address="575de">
<table name="Volts" address="61842"/>
</table>

<table name="MAF Scaling Part 2" address="57636">
<table name="Volts" address="6189a"/>
</table>

<table name="MAF Scaling Part 3" address="5768e">
<table name="Volts" address="618f2"/>
</table>

2009 Ralliart ROM 53600010


<table name="MAF Scaling Part 1" address="57586">
<table name="Volts" address="6164a"/>
</table>

<table name="MAF Scaling Part 2" address="575de">
<table name="Volts" address="616a2"/>
</table>

<table name="MAF Scaling Part 3" address="57636">
<table name="Volts" address="616fa"/>
</table>

Those are the one's I've tested / worked on.

Thanks Golden! Much appreciated.

Do you have the evo10base for those? Just want to confirm how many elements are in each part. I may have to implement an adjustable number of elements or something. I imagine not everyone wants to use the scaling in the spreadsheet, especially those who have already started using another definition.


Here is what you will need if you want more in there:
2010 Evo X GSR ROM 55570005
And
2010 Ralliart ROM 55590006


<table name="MAF Scaling Part 1" address="575de">
<table name="Volts" address="61842"/>
</table>

<table name="MAF Scaling Part 2" address="57636">
<table name="Volts" address="6189a"/>
</table>

<table name="MAF Scaling Part 3" address="5768e">
<table name="Volts" address="618f2"/>
</table>

2009 Ralliart ROM 53600010


<table name="MAF Scaling Part 1" address="57586">
<table name="Volts" address="6164a"/>
</table>

<table name="MAF Scaling Part 2" address="575de">
<table name="Volts" address="616a2"/>
</table>

<table name="MAF Scaling Part 3" address="57636">
<table name="Volts" address="616fa"/>
</table>

Those are the one's I've tested / worked on.

I double checked both, and the values you have in your spread sheet are correct for the tables I listed. I made sure all 3 columns started and ended at the same values / voltages.

Edit: they all use the same evo10base in my setup.

xprimnt ,

How would I go about getting the LTFT_Current in evo scan? Im at work so I don't have it in front of me.

Is there a way to look up MUT50? and have the address logged?

In EcuFlash, you open your MUT Table and look up MUT50. That is the value you put in EvoScan to log.

Example:

55570005 Rom
http://i746.photobucket.com/albums/xx108/EvoX_2010/Misc/MUT.jpg

I would add a "23" to the front of "8088C1" and enter an address of "238088C1" in EvoScan. Equation is the same as the other trims.

In EcuFlash, you open your MUT Table and look up MUT50. That is the value you put in EvoScan to log.

Example:

55570005 Rom
http://i746.photobucket.com/albums/xx108/EvoX_2010/Misc/MUT.jpg

I would add a "23" to the front of "8088C1" and enter an address of "238088C1" in EvoScan. Equation is the same as the other trims.

awesome !! thanks for the quick HOW-TO

and to confirm MUT50 is for 2008 GSR or 2010 GSR ??

Exactly as Golden said (and illustrated nicely).

What ROMID Storm?

EDIT: MUT50 is for everything. The address might change between ROMs, but MUT50 stays the same ;)

It's for all maps; thats why we post it as MUT50, so you can look it up in your map.


http://evoecu.logic.net/wiki/MUT_Requests <- Some work, some don't.

Exactly as Golden said (and illustrated nicely).

What ROMID Storm?

EDIT: MUT50 is for everything. The address might change between ROMs, but MUT50 stays the same ;)

ahhh I see great to know :D

thanks again, and I have the same inherit as you 52680015.xml.

Sorry Golden, I'm going to beat this to death.

I just want to make sure before I add it into the spreadsheet. 44-44-42 elements? Scaling type the same?

evo10base.xml

<table name="MAF Scaling Part 1" category="Fuel" type="2D" level="2" scaling="uint16">
<table name="Volts" type="Y Axis" elements="44" scaling="VoltsADC1023"/>
</table>

<table name="MAF Scaling Part 2" category="Fuel" type="2D" level="2" scaling="uint16">
<table name="Volts" type="Y Axis" elements="44" scaling="VoltsADC1023"/>
</table>

<table name="MAF Scaling Part 3" category="Fuel" type="2D" level="2" scaling="uint16">
<table name="Volts" type="Y Axis" elements="42" scaling="VoltsADC1023"/>
</table>

It's your thread. :D

Identical:

<table name="MAF Scaling Part 1" category="Fuel" type="2D" level="2" scaling="uint16">
<table name="Volts" type="Y Axis" elements="44" scaling="VoltsADC1023"/>
</table>

<table name="MAF Scaling Part 2" category="Fuel" type="2D" level="2" scaling="uint16">
<table name="Volts" type="Y Axis" elements="44" scaling="VoltsADC1023"/>
</table>

<table name="MAF Scaling Part 3" category="Fuel" type="2D" level="2" scaling="uint16">
<table name="Volts" type="Y Axis" elements="42" scaling="VoltsADC1023"/>
</table>

I got that part from Clipse's thread on scaling the MAF. I had to change the addresses around in my table to make it match his definitions, because the definitions that I found were different.
I wanted my tables to match the definitions of everyone else's on purpose.

If you want, I can send you roms and definitions if your OCD isn't subdued by this post. :innocent:

If you want, I can send you roms and definitions if your OCD isn't subdued by this post. :innocent:

Ha, thanks :thumbup:

Appreciated. No need for further feeding my OCD.

Thanks to Golden, I've added a few more definitions, etc to the instructions. No calculations have changed.

Hey xprimnt, i notice in your base definition, this was not included..

<scaling name="MAFVoltScaling" units="units" toexpr="x*5/1024" frexpr="x*1024/5" format="%.3f" min="0" max="5" inc="1" storagetype="uint16" endian="big"/>


should I remove that from my base when pasting in your definition?

ECUFlash or EvoScan?

no its for ecu flash ( i think) It is from clipse's how-to on maf scaling, he has the 52680015.xml then the evo10base.xml and that is listed at the top of the base.xml


So i am assuming you do not have that in your roms?

Ah, I remember.

I forgot that I had done that before. I changed the scaling unit to one that already exists, so it is no longer required. No harm in leaving it in, however.

Ah, I remember.

I forgot that I had done that before. I changed the scaling unit to one that already exists, so it is no longer required. No harm in leaving it in, however.

okay thanks for the heads up :)

EDIT:

Pasted the definitions in, i like how the are vertical columns now :D. How ever, I am only showing the values, not volts? Is that how you have it set up?

Sorry for the 9832234 questions :-/

double posttt

okay thanks for the heads up :)

EDIT:

Pasted the definitions in, i like how the are vertical columns now :D. How ever, I am only showing the values, not volts? Is that how you have it set up?

Sorry for the 9832234 questions :-/

No, I have it nicely set up

Volts MAF
Volt MAF
Volt MAF

Etc.

I copied out of my definitions, I haven't actually tried starting from scratch. If it doesn't work well, I may have to see what's up. You have both the 15.xml and evo10base.xml pasted in?

Do you have this line in you evo10base?

<scaling name="VoltsADC1023" units="Volts" toexpr="x*5/1023" frexpr="x*1023/5" format="%.2f" min="0" max="5" inc="0.05" storagetype="uint16" endian="big"/>


Anyone else run into problems?

No, I have it nicely set up

Volts MAF
Volt MAF
Volt MAF

Etc.

I copied out of my definitions, I haven't actually tried starting from scratch. If it doesn't work well, I may have to see what's up. You have both the 15.xml and evo10base.xml pasted in?

Anyone else run into problems?

correct, pasted in both files, in my modified folder and the evo folder, have my backup on my desktop so it doesnt get confused.

attached is a screen shot so you can see what im looking at

Oh, I see what you are asking.

Those values are Volts * 100. So if you divide by 100, you would get volts. It's kind of like leaving the decimal off.

Edit: yes his spread sheet uses numbers like the ones in your scalings. Here is a quick sample:
3.87 17151
3.91 17780
3.95 18422
3.99 19079
4.03 19750
4.07 20429
4.11 21119
4.14 21830
4.18 22552
4.22 23287
4.26 24034
4.30 24780
4.34 25542
4.38 26329
4.42 27145
4.46 27983
4.50 28847
4.54 29700
4.57 30553

No, I have it nicely set up

Volts MAF
Volt MAF
Volt MAF

Etc.

I copied out of my definitions, I haven't actually tried starting from scratch. If it doesn't work well, I may have to see what's up. You have both the 15.xml and evo10base.xml pasted in?

Do you have this line in you evo10base?

<scaling name="VoltsADC1023" units="Volts" toexpr="x*5/1023" frexpr="x*1023/5" format="%.2f" min="0" max="5" inc="0.05" storagetype="uint16" endian="big"/>

Anyone else run into problems?


Fixed with the code you inserted
looks pretty now :)

also just added LTFT Current to evoscan, sooo now all i need to do is log then paste in my values and see what pops out :)

xPRimNT,
I just noticed that your "Paste into ATR" section needs all the values / 100. Cobb values look like this:
http://i746.photobucket.com/albums/xx108/EvoX_2010/Misc/CobbMAF.jpg


Edit: you may also want to have an area for them to paste in their raw values. (horizontal and then fill your section with (Cobb value * 100) )

Storm, I'll check when I get home how I have it set up. I thought I had copied and pasted, but I might have some slight differences that I haven't posted up. I'm on a TephraMOD ROM, so the def file is slightly different.


Golden, that's actually weird. I "half installed" (I don't have an AP) ATR, and opened up the MAF tables to see how they were set up. I copied it from there, and it didn't have decimals! I'll reconfirm. If it doesn't, then something is screwy with my install ;)

I wish I had my (tuning) computer here. I changed my stuff from G/s because of how my ATR was set up. I would rather have it in an actual unit like G/s.

Whatever, I'm switching over to gram/s, I'll post the new stuff when I get home. Sorry Storm, you'll have some changes to make ;)

I had the MAF values originally as uint16, basically the raw values. I'll switch to grams/s

Whatever, I'm switching over to gram/s, I'll post the new stuff when I get home. Sorry Storm, you'll have some changes to make ;)

I had the MAF values originally as uint16, basically the raw values. I'll switch to grams/s


awww phooooy haha

its all good, all for the sake of development! im willing to hang in there, keep us posted ;)

I changed my mind.

Seems like the Cobb g/s is "g/s" in quotes, not actual measurement of airflow. It seems that they just take the raw value (x) and divide it by 100. Grams per second would be slightly different. That doesn't really matter though, since we're dealing with percents and multiplicative changes to an existing value.

Instead, I simply made the spreadsheet more ATR friendly. No changes to the ECUFlash side of things (hopefully, let me know if it doesn't work).

entered all my logs and all the calculated values show up as '#DIV/0!' screen shot attached....

entered all my logs and all the calculated values show up as '#DIV/0!' screen shot attached....

I just used it today, so it should work!


Did you Cut and Paste, or Copy and Paste? For some reason, Cutting takes out the references to cells, so only copying really works.

Feel free to email me (ForumName at Gmaildotcom) if it isn't working.


(Start with a new copy, the references are already shot in that one ;) )

So, this algorythm assumes that you have perfectly dialed in injectors, right? And if you change both, injectors and intake, you can't really use this, can you?

Yes, you should be on stock injectors to do this with a new intake. Then switch over to new injectors and dial those in.

+1 on that. Change one thing at a time. Never add injectors and intake at the same time.

I just used it today, so it should work!


Did you Cut and Paste, or Copy and Paste? For some reason, Cutting takes out the references to cells, so only copying really works.

Feel free to email me (ForumName at Gmaildotcom) if it isn't working.


(Start with a new copy, the references are already shot in that one ;) )

Thats what I did!

I copied my values, then pasted them in.... ?

could it be the fact that I am not running 2007+ excel?

i am running pre xlsx

Thats what I did!

I copied my values, then pasted them in.... ?

could it be the fact that I am not running 2007+ excel?

i am running pre xlsx

I'll try and make it compatible tonight to see if that helps. Do the values work before you paste your log in? That is, no REF errors?

Good work guys! I just seen this thread! I'll keep an eye on this. Sorry won't be able help at all on this one. Don't have time right now to drive around with moving and all. Very nice work though!

I'll try and make it compatible tonight to see if that helps. Do the values work before you paste your log in? That is, no REF errors?


Im not sure, i remember when i first opened it last weekend on my computer and went to the second tab where the data analysis is, I clicked on the columns that are highlighted, i.e. LTFT_cruise etc....

and all of a sudden the values that are calculated did the same thing that happened to me.

so if its working for you, i think it may be a compatibility issue.

but I will try again this afternoon. start from scratch. or download xlsx and try it that way :D

great job there!

Ok Storm, try the compatible one.

Hopefully it works. I tested the xlsx one on another Windows computer and it worked fine, so I'm not sure what's up with yours.

Ok Storm, try the compatible one.

Hopefully it works. I tested the xlsx one on another Windows computer and it worked fine, so I'm not sure what's up with yours.


I have no clue what could be wrong.

copy and paste in the values again, and I have the same problem as last time.

I am running office 2003 for mac. could that be the issue?

I even tried to manually enter the values and the same thing happens.

I think it has to do with my machine not liking the formulas.

blahhh I dont know what to do?!?!?!

or it can be my data? I don't know.

or it can be my data? I don't know.

Send me your data, let me see if I can get it to work.

Honestly, on every computer I've tried, I just copy the columns from my log and paste it into the columns of the spreadsheet. It works fine every time ;)

yeah i dunno, I must be excel inept haha

having issues attaching files to the forum right now, im sending you an email!

Ok Storm, try the compatible one.

Hopefully it works. I tested the xlsx one on another Windows computer and it worked fine, so I'm not sure what's up with yours.


hey mate, just got it, works well in my mac.

do you have definitions for 53050006 and 53050009 ?

thx :)

Send me your data, let me see if I can get it to work.

Honestly, on every computer I've tried, I just copy the columns from my log and paste it into the columns of the spreadsheet. It works fine every time ;)


so I went over to my parents house and used my dads computer (running office 2007) and......


WORKED like a charm :)

After two cycles my trims have come down very nicely. my cruise trim is doing well, and my idle trim stayed at zero for most my drive to work going down to about -3 by the time I pulled into my parking space.

Yesterday it made it to -10 with in 10 miles of driving :confused2:

either way, now I can start focusing on dialing in my AFRs to finish up my tune :)

hey mate, just got it, works well in my mac.

do you have definitions for 53050006 and 53050009 ?

thx :)

I don't, sorry! I'm sure they're around somewhere though.

so I went over to my parents house and used my dads computer (running office 2007) and......


WORKED like a charm :)

After two cycles my trims have come down very nicely. my cruise trim is doing well, and my idle trim stayed at zero for most my drive to work going down to about -3 by the time I pulled into my parking space.

Yesterday it made it to -10 with in 10 miles of driving :confused2:

either way, now I can start focusing on dialing in my AFRs to finish up my tune :)

Glad to hear it's working. Sorry I can't get it more compatible, not really sure what's up. Just do a few more until it gets dialed in.

got the xml's xprimnt. what to do next LoL ....

The MAF scaling one has some instructions in it. Basically, log what it says to log and paste it in.

The other one is a bit more difficult so I haven't written up anything. Just play with it! Log what you think you want to log for a pull and make it work for you.

k more testing time & will report back hehehe ... i'm testing newly added features on the edm too :godance:

Thanks for the sheet, xprimnt :) It's been a huge help!

I have a question based on an observation. I know that you're sheet is based on Hollywood's MAF scaling method, which applies observed trim deltas up to 2.70v, then calculates the corresponding trims at higher voltages.

When I graph my results, I'd expect a nice smooth parabolic increase, but I always get a strange artifact right at 2.7v. When my trims shift my graph upwards until 2.7v, then at 2.7v I apply an increment on the OLD (lower) value, my 2.7v value is lower than the previous value. Since it's exponential, it catches up over time. While it's not a major shift, I assume that this will cascade and be a 'bigger deal' at the higher voltages, since its exponential. I've noticed this through my own testing, and looking at some data given to me by others.

I think the problem is that, after calculating the average change (P44), we're applying that change to the OLD data. I think what we should be doing is applying that change to the last calculated cell, then from then, applying the next change the the previous cell. Basically throwing out the old data after 2.7v. To pull this off well, we'll have to change everything else a bit since apply a static multiplier to a exponential set is a bad idea (otherwise linear scaling would work).

Basically, I'm working on revamping the sheet to use a Exponential Weighted Moving Average (http://en.wikipedia.org/wiki/Moving_average#Exponential_moving_average). I'm not sure if I can do this in excel, but I'll try and let you know what I come up with :)

Before spending too much time on it, does this seem like a sound method, or am I just asking for trouble?

You're absolutely right.

I think that the weight exp average might be a decent solution, and I'm more than willing to work it in.

Hit me with an email at any time (Forumname atgmail dotcom) and we'll figure something out.

Thanks for contributing! I know that the Hollywood/Cobb way isn't perfect, but it was what I was able to get good results with. I definitely manually smoothed things out a few times at the OL change point, so if this works you get s thumbs up :)

Bump - is there an updated version of this spreadsheet?

I reworked the sheet quite a bit. It worked very well after 3 runs (trims less than +-0.5). My version is here:

http://dl.dropbox.com/u/386788/blk_MAF_Scale.xlsx

Edit: some instructions

Use the 'Scaling Helper' tab to go from your current values to new baseline values based on your current scaling tables, current MAF housing size, and new MAF housing size. it uses a linear transform, which is NOT accurate, but gets close enough to do some test logs.

After you've logged, use the Analysis tab. Paste your MAF tables in the Logged Mass column. Then go to the Log Data tab and supply your logs in columns A-C. Go back to the Analysis tab and get your new scaling table in the New Mass column. Make sure it looks sane, then paste it in EcuFlash, save, reflash.

Repeat this 3 times and you should be good to go :)

I've modified mine some too. The big thing is to make sure you are not applying the trims when STFT = 0, because you are in open loop.

bumping this up for those that were asking

thank you 2NR

Don't thank me... thank the OP! This spreadsheet saved my ass.

Thanks for the feedback everyone.

I haven't had time to work on these sheets in a LONG time, I barely have the chance to log in these days. Hopefully things will slow down a bit at some point, but for now... Sigh.


The further contributions are awesome blk, thanks for posting them up. I'll be sure to test them at some point, but the details look good.

As for everyone trying these out - Please feel free to adjust things as you see fit, and share them with the community. It would be awesome to see more detailed yet easy to use tuning spreadsheets, so keep it up!

I have a few questions. I used to blk-majik's sheet to scale my MAF and it seems to have worked pretty good for the LTFT cruise (sitting at about -0.98 right now). But my LTFT idle is a bit farther out than I would like, at 5.08.

I guess I'm not entirely sure if what I did was correct, so let me explain and please let me know if I did something wrong. I disconnected the battery to reset the fuel trims, used the sheet and the "scaling helper" as my starting point (the first time I tried to scale the maf I didn't use the scaling helper and curve didn't look right...basically above to 2.66ish area was about stock...so I started over using it). Then I drove around for about 20 mins while logging and plugged those values into the sheet along with the maf scale used during that particular log. Although I used the LTFT Current as opposed to Cruise.

I did not reset the fuel trims/unplugg the battery in between each drives (only before starting the whole process), I'm not sure if I was suppose to or not. Also I wasn't exactly sure what type of driving I was suppose to do, but it was mostly a little highway/roads without stop lights, with a little stop and go mixed in. Light, no fast pulls or anything like that.

So did I do something wrong, or do I just need to do some more logging and adjusting (I've done 3 logs/adjustments so far plus the initial scaling helper)?