Help scaling MAF
Help scaling MAF
Well I spent a couple hours reading and I havent seen a full answer on how to scale your MAF correctly. I usually don't post because I like to figure stuff out on my own, but this is driving me nuts. I added an ETS intake and my car idles like complete crap. I remember that some intakes do this and some don't and the MAF needs to be scaled if it idles funny. I read the following threads on MAF scaling:
https://www.evolutionm.net/forums/sh...=rescaling+maf
https://www.evolutionm.net/forums/sh...d.php?t=240323
And the one by SophieSleeps that has no pictures. I also read the Excel spreadsheet that l2r99gst made about it.
What I have read and think have taken from the three threads and the spreadsheet is that you need to log your STFT and LTFT as well as Hz to see what cells to modify to smooth your idle. I read a post by SophieSleeps that you add 140 to whatever cell then multiply by 10%, whatever you get from that you add to the cell you want to modify. Seems like there are two loggers being used - EvoScan and Mitsulogger. I dont want to post my entire logs but here is the gist of both.
EvoScan
LTFT - 100 through the entire log
STFT (o2feedbacktrim) - 110-115 mainly with the occasional spike up to 125. Occasional drop to 89 when AFRs go lean at idle.
AirFlow - 31.45 pretty much all the way through. There are 3 different instances where it is as high as 37 and as low as 25.
Mitsulogger
Looking at the values it seems to already read in %. Please correct me if I am wrong.
LTFT - 0
MTFT - -2
STFT - 4
AirFlow - Mainly 37.74. Upwards to 44.03 for like 2 seconds. Then it goes to 31.45 for a bit.
Taking a look at my ROM for MAF scaling here are the first three entries:
19 | 145
25 | 152
50 | 165
I know that these are the values I will need to edit from the AirFlow log. Now going by SophieSleeps' post, I take my original data of 145, add 140. I get 285. 10% of 285 is 28.5. 145+28.5 = 173.5. Now if I do that to the first three cells, I should have 173.5, 181.2, 195.5. I am just wondering if this is correct or am I missing something that I might have looked over?
https://www.evolutionm.net/forums/sh...=rescaling+maf
https://www.evolutionm.net/forums/sh...d.php?t=240323
And the one by SophieSleeps that has no pictures. I also read the Excel spreadsheet that l2r99gst made about it.
What I have read and think have taken from the three threads and the spreadsheet is that you need to log your STFT and LTFT as well as Hz to see what cells to modify to smooth your idle. I read a post by SophieSleeps that you add 140 to whatever cell then multiply by 10%, whatever you get from that you add to the cell you want to modify. Seems like there are two loggers being used - EvoScan and Mitsulogger. I dont want to post my entire logs but here is the gist of both.
EvoScan
LTFT - 100 through the entire log
STFT (o2feedbacktrim) - 110-115 mainly with the occasional spike up to 125. Occasional drop to 89 when AFRs go lean at idle.
AirFlow - 31.45 pretty much all the way through. There are 3 different instances where it is as high as 37 and as low as 25.
Mitsulogger
Looking at the values it seems to already read in %. Please correct me if I am wrong.
LTFT - 0
MTFT - -2
STFT - 4
AirFlow - Mainly 37.74. Upwards to 44.03 for like 2 seconds. Then it goes to 31.45 for a bit.
Taking a look at my ROM for MAF scaling here are the first three entries:
19 | 145
25 | 152
50 | 165
I know that these are the values I will need to edit from the AirFlow log. Now going by SophieSleeps' post, I take my original data of 145, add 140. I get 285. 10% of 285 is 28.5. 145+28.5 = 173.5. Now if I do that to the first three cells, I should have 173.5, 181.2, 195.5. I am just wondering if this is correct or am I missing something that I might have looked over?
I havent changed any of my MAF settings yet, but I want to do that ASAP. I just installed ECUFlash on Parallels so I can get a quick screenshot of my MAF scale, otherwise I would have to restart into Windows. I havent changed the load over to Hz yet. I think I see where I went wrong in reading my logs but I am not too sure. My main question is how do you get the percentage number to change your MAF scale table? I will attach a screenshot of my logs as well.
MAF scale table
EvoScan Log
Mitsulogger Log
MAF scale table
EvoScan Log
Mitsulogger Log
I am not at liberty to share my values as my tuner worked a hell of a long time on it. I can tell you though that every value in the MAF scale table WILL be a lot lower than the stock values.
Keep plugging away at it and I'll tell you if you're headed in the right direction...
-E
Keep plugging away at it and I'll tell you if you're headed in the right direction...
-E
Trending Topics
Well, it wouldn't take so much anymore E with the new software that's been developed. I'd imagine I could get it done in 2-3 revisions now.
Thank you for keeping my maps off the forum E!
Thank you for keeping my maps off the forum E!
I dont see why "scaling my MAF asap" is not possible. Regardless of how long it takes, it is as soon as possible. I am just trying to scale my idle as of now, so it doesnt die on idle.
Back on topic, you said that MAF values will be lower than stock. Every value I have seen is slightly higher than stock. I guess I am not headed in the right direction.
Back on topic, you said that MAF values will be lower than stock. Every value I have seen is slightly higher than stock. I guess I am not headed in the right direction.
OK, let me clear a few things up for you:
1. The percentage of correction that each cell in your MAF scaling table may be different. You don't just find a correction, like 10%, and apply it to all cells.
2. The way that you find the percentage correction that you need is to log what you have logged already and to add the long term fuel trim correction and short term fuel trim correction at the logged Hz value (LTFT Lo at idle and LTFT mid at cruise). Then, that % correction is the % correction that cell in the MAF scaling table needs (after adding 140 to the value first).
Also, another huge note....make sure that whatever logger your are using has the correct long term and short terms fuel trim formulas and requests. I believe Mitsulogger has them correct by default.
The request are:
LTFT Lo - 0C
LTFT Mid - 0D
LTFT Hi - 0E
STFT - 0F
The formula is .1961x-25 (may not be that exact format, but that is the formula).
Looking at your numbers, you don't need your MAF scaled, at least not at the Hz values for the logs that you have posted. Everything looks fine to me.
Your idle only needs +4% correction, which is good enough not to mess with.
What you should do now is log a steady cruise at varying Hz levels and see what your LTFT mid and STFT are. This will tell you if you need any correction at higher Hz levels.
If you do need any kind of correction, the percentage is found by simply adding the LTFT and STFT (LTFT Mid for cruise, LTFT Lo for idle). So, for example, let's say that you were cruising at 125Hz (steady RPM), and your LTFT Mid was -10% and your STFT was bouncing around 0. Then, you would need a -10% correction.
So, you would take the value at your 125 Hz row in your MAF scale, add 140 to it, and find 10% of that number. Subtract that resulting number from the value in the MAF scaling table. You would have to do this for a few Hz numbers to see what change you need for the various cells, or simply taper it as you see fit.
Note that at higher Hz values, you will most likely be transitiioning to open loop, so you won't necessarily have to use your MAF scaling table anymore. You can simply use your open loop fuel maps for compensation at that point.
Eric
1. The percentage of correction that each cell in your MAF scaling table may be different. You don't just find a correction, like 10%, and apply it to all cells.
2. The way that you find the percentage correction that you need is to log what you have logged already and to add the long term fuel trim correction and short term fuel trim correction at the logged Hz value (LTFT Lo at idle and LTFT mid at cruise). Then, that % correction is the % correction that cell in the MAF scaling table needs (after adding 140 to the value first).
Also, another huge note....make sure that whatever logger your are using has the correct long term and short terms fuel trim formulas and requests. I believe Mitsulogger has them correct by default.
The request are:
LTFT Lo - 0C
LTFT Mid - 0D
LTFT Hi - 0E
STFT - 0F
The formula is .1961x-25 (may not be that exact format, but that is the formula).
Looking at your numbers, you don't need your MAF scaled, at least not at the Hz values for the logs that you have posted. Everything looks fine to me.
Your idle only needs +4% correction, which is good enough not to mess with.
What you should do now is log a steady cruise at varying Hz levels and see what your LTFT mid and STFT are. This will tell you if you need any correction at higher Hz levels.
If you do need any kind of correction, the percentage is found by simply adding the LTFT and STFT (LTFT Mid for cruise, LTFT Lo for idle). So, for example, let's say that you were cruising at 125Hz (steady RPM), and your LTFT Mid was -10% and your STFT was bouncing around 0. Then, you would need a -10% correction.
So, you would take the value at your 125 Hz row in your MAF scale, add 140 to it, and find 10% of that number. Subtract that resulting number from the value in the MAF scaling table. You would have to do this for a few Hz numbers to see what change you need for the various cells, or simply taper it as you see fit.
Note that at higher Hz values, you will most likely be transitiioning to open loop, so you won't necessarily have to use your MAF scaling table anymore. You can simply use your open loop fuel maps for compensation at that point.
Eric
Last edited by l2r99gst; Jun 10, 2008 at 07:05 AM.
First, thanks for the reply Eric. I read your Excel spreadsheet and your calculations and formulas just blew right over my head. Second, it will take me a bit to grasp all of this since I am still waking up. Maybe I should have some coffee.
Let me make sure I understand this correctly. Add my STFT and LTFT to get the % value of what I need to do. In this case, 4% at idle. First value of my MAF scale is 145. 145 + 140 = 245 * .04 = 9.8. So I add 9.8 to my first value, and go down the list according to the Hz I am logging.
Will I need to do corrections at idle and cruising to get my car to drive the way it used to before the intake?
Let me make sure I understand this correctly. Add my STFT and LTFT to get the % value of what I need to do. In this case, 4% at idle. First value of my MAF scale is 145. 145 + 140 = 245 * .04 = 9.8. So I add 9.8 to my first value, and go down the list according to the Hz I am logging.
Will I need to do corrections at idle and cruising to get my car to drive the way it used to before the intake?
There is another thing you can try. If your Low (idle) and Mid (cruise) LTFT's are in fact 0 and -2, that's not something you'd want to fix. It's almost perfect. Maybe you haven't logged enough after changing the intake though for them to adjust more. But it doesnt take more than 5 min of idling usually.
What sometimes happens with intakes is the airflow gets stalled in them at low rpm or MAF reads it incorrectly, so one easy solution you can try is bump the idle a bit. Maybe to 900 or 950 or 1000 and see if it helps.
Also finish rescaling the numbers. The left column isn't done yet. It'll be easier to figure it out when the numbers on the left will represent Hz that you log with Mitsulogger or Evoscan.
Your table:
What sometimes happens with intakes is the airflow gets stalled in them at low rpm or MAF reads it incorrectly, so one easy solution you can try is bump the idle a bit. Maybe to 900 or 950 or 1000 and see if it helps.
Also finish rescaling the numbers. The left column isn't done yet. It'll be easier to figure it out when the numbers on the left will represent Hz that you log with Mitsulogger or Evoscan.
Your table:
Last edited by mplspilot; Jun 10, 2008 at 08:08 AM.
OK, let me clear a few things up for you:
1. The percentage of correction that each cell in your MAF scaling table may be different. You don't just find a correction, like 10%, and apply it to all cells.
2. The way that you find the percentage correction that you need is to log what you have logged already and to add the long term fuel trim correction and short term fuel trim correction at the logged Hz value (LTFT Lo at idle and LTFT mid at cruise). Then, that % correction is the % correction that cell in the MAF scaling table needs (after adding 140 to the value first).
Also, another huge note....make sure that whatever logger your are using has the correct long term and short terms fuel trim formulas and requests. I believe Mitsulogger has them correct by default.
The request are:
LTFT Lo - 0C
LTFT Mid - 0D
LTFT Hi - 0E
STFT - 0F
The formula is .1961x-25 (may not be that exact format, but that is the formula).
Looking at your numbers, you don't need your MAF scaled, at least not at the Hz values for the logs that you have posted. Everything looks fine to me.
Your idle only needs +4% correction, which is good enough not to mess with.
What you should do now is log a steady cruise at varying Hz levels and see what your LTFT mid and STFT are. This will tell you if you need any correction at higher Hz levels.
If you do need any kind of correction, the percentage is found by simply adding the LTFT and STFT (LTFT Mid for cruise, LTFT Lo for idle). So, for example, let's say that you were cruising at 125Hz (steady RPM), and your LTFT Mid was -10% and your STFT was bouncing around 0. Then, you would need a -10% correction.
So, you would take the value at your 125 Hz row in your MAF scale, add 140 to it, and find 10% of that number. Subtract that resulting number from the value in the MAF scaling table. You would have to do this for a few Hz numbers to see what change you need for the various cells, or simply taper it as you see fit.
Note that at higher Hz values, you will most likely be transitiioning to open loop, so you won't necessarily have to use your MAF scaling table anymore. You can simply use your open loop fuel maps for compensation at that point.
Eric
1. The percentage of correction that each cell in your MAF scaling table may be different. You don't just find a correction, like 10%, and apply it to all cells.
2. The way that you find the percentage correction that you need is to log what you have logged already and to add the long term fuel trim correction and short term fuel trim correction at the logged Hz value (LTFT Lo at idle and LTFT mid at cruise). Then, that % correction is the % correction that cell in the MAF scaling table needs (after adding 140 to the value first).
Also, another huge note....make sure that whatever logger your are using has the correct long term and short terms fuel trim formulas and requests. I believe Mitsulogger has them correct by default.
The request are:
LTFT Lo - 0C
LTFT Mid - 0D
LTFT Hi - 0E
STFT - 0F
The formula is .1961x-25 (may not be that exact format, but that is the formula).
Looking at your numbers, you don't need your MAF scaled, at least not at the Hz values for the logs that you have posted. Everything looks fine to me.
Your idle only needs +4% correction, which is good enough not to mess with.
What you should do now is log a steady cruise at varying Hz levels and see what your LTFT mid and STFT are. This will tell you if you need any correction at higher Hz levels.
If you do need any kind of correction, the percentage is found by simply adding the LTFT and STFT (LTFT Mid for cruise, LTFT Lo for idle). So, for example, let's say that you were cruising at 125Hz (steady RPM), and your LTFT Mid was -10% and your STFT was bouncing around 0. Then, you would need a -10% correction.
So, you would take the value at your 125 Hz row in your MAF scale, add 140 to it, and find 10% of that number. Subtract that resulting number from the value in the MAF scaling table. You would have to do this for a few Hz numbers to see what change you need for the various cells, or simply taper it as you see fit.
Note that at higher Hz values, you will most likely be transitiioning to open loop, so you won't necessarily have to use your MAF scaling table anymore. You can simply use your open loop fuel maps for compensation at that point.
Eric
As in for idle, if you idle at 50Hz and Low LTFT is -5 you would take the number from 50Hz (152 or whatever) add 140 subtract 5% then reduce it by 140 and put it in the table.
Do the same for cruise and smooth the table.
I mean STFT bounces all over most of the time, how would you add it?
Am i missing something here?
To really tune the entire table you have to do what l2r stated ... look at both the LTFT and the STFT. This is because the LTFT is an average based on many of the hz values. If you can hold a certain hz value the STFT will reflect how much correction is needed at that specific airflow value beyond the average LTFT.