I've had an idea
I've had an idea
I was recently thinking about my tephra rom with map switching and running E85,
there are two different injector scalings for the two different fuels and this scaling is derived from the fuel trims,
when you are tuning for E85 you tune the injector scaling till you get the fuel trims to below +-5% and then tune your fuel maps,
we all know the trims are there to try and help the ecu reach the target AFR set in the fuel maps,
if your injector scalings are way off the trims follow to try and keep the AFR close to the maps
what I noticed is when using a mixture of E85 and 95 ron the trims can go way off as it tries to struggle to trim for the two different fuels mixed because the injector scaling is static,
what would happen if we patched the rom so the fuel trims changed the Injector scaling automatically
you would never have to worry about your trims being off again
and for the purpose of mixing E85 with normal 95 ron the ecu would change the scaling to keep the trims below the 5% mark and you could eliminate the alt maps injector scaling as the main scaling would change to suit the fuel mix and ensure the AFR was always in trim,
now I know that this is the opposite of what the trims do but the trims are limited to 20% and my scalings are 496 for E85 on 1000cc injectors
and 713 for normal 95 ron
and this span is too much for the trims to compensate for,
there are two different injector scalings for the two different fuels and this scaling is derived from the fuel trims,
when you are tuning for E85 you tune the injector scaling till you get the fuel trims to below +-5% and then tune your fuel maps,
we all know the trims are there to try and help the ecu reach the target AFR set in the fuel maps,
if your injector scalings are way off the trims follow to try and keep the AFR close to the maps
what I noticed is when using a mixture of E85 and 95 ron the trims can go way off as it tries to struggle to trim for the two different fuels mixed because the injector scaling is static,
what would happen if we patched the rom so the fuel trims changed the Injector scaling automatically
you would never have to worry about your trims being off again
and for the purpose of mixing E85 with normal 95 ron the ecu would change the scaling to keep the trims below the 5% mark and you could eliminate the alt maps injector scaling as the main scaling would change to suit the fuel mix and ensure the AFR was always in trim,
now I know that this is the opposite of what the trims do but the trims are limited to 20% and my scalings are 496 for E85 on 1000cc injectors
and 713 for normal 95 ron
and this span is too much for the trims to compensate for,
Evolved Member
Joined: May 2006
Posts: 2,399
Likes: 6
From: UK
If you have the injector scaling in RAM, you could add a small piece of code to the main loop to shift the injector scaling based on one or an average of 2 or 3 of the LTFT, this would tend to neutralise the LTFT and move the injector size instead.
it would mean you could have a true flexifuel car as the trims would rescale the injectors depending on the fuel mix??
I have a wideband displayed on my mxl dash could test it see how it works
if it is possible to mix the fuel and if the ecu could compensate for the various mixes???
Barry I can see a few problems.
1. The stock ecu doesn't rely on the stock fuel trims for WOT fueling for a reason, they drift as the car is warming up/down. The fuel trims are simply a way to provide feedback at low loads. To do WOT "compensation" as you suggest one should use a wideband O2 sensor and do feedback at WOT. I've done it on a MAFT setup and it works well. That is until the wideband starts reading wrong and you go lean because of it.
2. What happens when your narrowband sensor starts crapping out? I've had it happen even when the hard upper and lower limits are intact on the ecu. The car goes extremely lean and wants to die. The only way to keep the car running is to go WOT (going into open loop) to provide the fueling to keep the car moving to your destination till you can fix the issue.
3. You are assuming that all the calibration settings in the ecu (injector latency, airflow sensor calibration, humidity compensation charts, baro pressure compensation charts, injector linearity,etc....) are 100% correct. If they were spot on at low loads when the fuel trims were "adjusting" and 10% off when you are WOT then you'd have a 10% fueling error. You'll notice that 10% is a huge error. You'll also notice that in bone stock form the evo does in fact show approximately a 10% error. IE: if you put in 10:1 afr on your WOT tables the car runs about 11:1.
4. What if you are in draw through setup and you have a charge pipe leak. The airflow sensor sees less air during no boost driving, your trims "compensate" for this, and you go lean at WOT.
For awhile now I've wanted a potentiometer wired into an ecu input to change the injector scaling on the fly within certain hard limits that would be set up in the ecu. That way I can make a small adjustment, look over at my wideband, make another adjustment etc. until I'm good. When I do t E85 mixture I can usually adjust my global on the first and if not the second flash to get the exact same AFR.
1. The stock ecu doesn't rely on the stock fuel trims for WOT fueling for a reason, they drift as the car is warming up/down. The fuel trims are simply a way to provide feedback at low loads. To do WOT "compensation" as you suggest one should use a wideband O2 sensor and do feedback at WOT. I've done it on a MAFT setup and it works well. That is until the wideband starts reading wrong and you go lean because of it.
2. What happens when your narrowband sensor starts crapping out? I've had it happen even when the hard upper and lower limits are intact on the ecu. The car goes extremely lean and wants to die. The only way to keep the car running is to go WOT (going into open loop) to provide the fueling to keep the car moving to your destination till you can fix the issue.
3. You are assuming that all the calibration settings in the ecu (injector latency, airflow sensor calibration, humidity compensation charts, baro pressure compensation charts, injector linearity,etc....) are 100% correct. If they were spot on at low loads when the fuel trims were "adjusting" and 10% off when you are WOT then you'd have a 10% fueling error. You'll notice that 10% is a huge error. You'll also notice that in bone stock form the evo does in fact show approximately a 10% error. IE: if you put in 10:1 afr on your WOT tables the car runs about 11:1.
4. What if you are in draw through setup and you have a charge pipe leak. The airflow sensor sees less air during no boost driving, your trims "compensate" for this, and you go lean at WOT.
For awhile now I've wanted a potentiometer wired into an ecu input to change the injector scaling on the fly within certain hard limits that would be set up in the ecu. That way I can make a small adjustment, look over at my wideband, make another adjustment etc. until I'm good. When I do t E85 mixture I can usually adjust my global on the first and if not the second flash to get the exact same AFR.
Trending Topics
Barry I can see a few problems.
1. The stock ecu doesn't rely on the stock fuel trims for WOT fueling for a reason, they drift as the car is warming up/down. The fuel trims are simply a way to provide feedback at low loads. To do WOT "compensation" as you suggest one should use a wideband O2 sensor and do feedback at WOT. I've done it on a MAFT setup and it works well. That is until the wideband starts reading wrong and you go lean because of it.
2. What happens when your narrowband sensor starts crapping out? I've had it happen even when the hard upper and lower limits are intact on the ecu. The car goes extremely lean and wants to die. The only way to keep the car running is to go WOT (going into open loop) to provide the fueling to keep the car moving to your destination till you can fix the issue.
3. You are assuming that all the calibration settings in the ecu (injector latency, airflow sensor calibration, humidity compensation charts, baro pressure compensation charts, injector linearity,etc....) are 100% correct. If they were spot on at low loads when the fuel trims were "adjusting" and 10% off when you are WOT then you'd have a 10% fueling error. You'll notice that 10% is a huge error. You'll also notice that in bone stock form the evo does in fact show approximately a 10% error. IE: if you put in 10:1 afr on your WOT tables the car runs about 11:1.
4. What if you are in draw through setup and you have a charge pipe leak. The airflow sensor sees less air during no boost driving, your trims "compensate" for this, and you go lean at WOT.
For awhile now I've wanted a potentiometer wired into an ecu input to change the injector scaling on the fly within certain hard limits that would be set up in the ecu. That way I can make a small adjustment, look over at my wideband, make another adjustment etc. until I'm good. When I do t E85 mixture I can usually adjust my global on the first and if not the second flash to get the exact same AFR.
1. The stock ecu doesn't rely on the stock fuel trims for WOT fueling for a reason, they drift as the car is warming up/down. The fuel trims are simply a way to provide feedback at low loads. To do WOT "compensation" as you suggest one should use a wideband O2 sensor and do feedback at WOT. I've done it on a MAFT setup and it works well. That is until the wideband starts reading wrong and you go lean because of it.
2. What happens when your narrowband sensor starts crapping out? I've had it happen even when the hard upper and lower limits are intact on the ecu. The car goes extremely lean and wants to die. The only way to keep the car running is to go WOT (going into open loop) to provide the fueling to keep the car moving to your destination till you can fix the issue.
3. You are assuming that all the calibration settings in the ecu (injector latency, airflow sensor calibration, humidity compensation charts, baro pressure compensation charts, injector linearity,etc....) are 100% correct. If they were spot on at low loads when the fuel trims were "adjusting" and 10% off when you are WOT then you'd have a 10% fueling error. You'll notice that 10% is a huge error. You'll also notice that in bone stock form the evo does in fact show approximately a 10% error. IE: if you put in 10:1 afr on your WOT tables the car runs about 11:1.
4. What if you are in draw through setup and you have a charge pipe leak. The airflow sensor sees less air during no boost driving, your trims "compensate" for this, and you go lean at WOT.
For awhile now I've wanted a potentiometer wired into an ecu input to change the injector scaling on the fly within certain hard limits that would be set up in the ecu. That way I can make a small adjustment, look over at my wideband, make another adjustment etc. until I'm good. When I do t E85 mixture I can usually adjust my global on the first and if not the second flash to get the exact same AFR.
good stuff mate cheers for that but at least it might get us thinking about a solution to be able to adjust the scaling for flexifuel or if we could use an input with the ethanol sensor to adjust the scaling
Thread
Thread Starter
Forum
Replies
Last Post
Raceghost
ECU Flash
22
Jun 1, 2016 12:30 AM
nj1266
Evo X Engine Management / Tuning Forums
125
Jan 12, 2010 07:37 AM