ATTN: Tephra, or any other "developers"
Feb 28, 2010, 10:18 AM
#16
Newbie
Thread Starter
Join Date: Apr 2009
Location: Arvada, CO
Posts: 14
Likes: 0
Received 0 Likes
on
0 Posts
PharmEcis you seem to be confusing a MAP based car and MAF based car. Mitsu ecus are in no way similar to Honda ecus.
Mitsu MAF based cars only measure airflow, and have no idea what the vacuum/pressure is, most pre 95 cars don't even have a MDP sensor.
Load % which matches kPa pretty close is taken from the OBD2 section of code, so people could have easy to understand units for the table. Evos pre 7 don't even have Load % or OBD2.
Mitsu's unit of airflow is liters per revolution, often simplified to grams of air per revolution. The ecu injects 1 gram of fuel if there is 14.7 grams of air metered, or what ever the target AFR is, it doesn't care if manifold pressure is -20 inHg or 50 psi.
Mitsu MAF based cars only measure airflow, and have no idea what the vacuum/pressure is, most pre 95 cars don't even have a MDP sensor.
Load % which matches kPa pretty close is taken from the OBD2 section of code, so people could have easy to understand units for the table. Evos pre 7 don't even have Load % or OBD2.
Mitsu's unit of airflow is liters per revolution, often simplified to grams of air per revolution. The ecu injects 1 gram of fuel if there is 14.7 grams of air metered, or what ever the target AFR is, it doesn't care if manifold pressure is -20 inHg or 50 psi.
Is the Evo MAF or MAP based? Sorry if its a noob question, just trying to learn and [i]attempt[/i keeping up with trying to learn here lol
Feb 28, 2010, 10:39 AM
#17
Newbie
Join Date: Aug 2006
Location: Sheffield England
Posts: 30
Likes: 0
Received 0 Likes
on
0 Posts
They are MAF based, That is what ceddy is saying.
There is a mod to make it MAP based called speed density. This is what the BIG boost high power boys like to use but if your systems are anything like evo ones ( Very likely) even boosted engines will be able to run off just the MAF..
There is a mod to make it MAP based called speed density. This is what the BIG boost high power boys like to use but if your systems are anything like evo ones ( Very likely) even boosted engines will be able to run off just the MAF..
Feb 28, 2010, 06:49 PM
#18
Evolving Member
iTrader: (3)
Join Date: Jul 2003
Location: Silver Spring, MD
Posts: 256
Likes: 0
Received 0 Likes
on
0 Posts
You know what... I should just shut up and let you guys bash your heads in trying to figure this out.
After all, I haven't been playing w/ a tactrix cable and 3G ECU's or ROMs since they first came out. I haven't tried rescaling maps beyond 100% load to see if I could actually get the ECU to read higher into the table.
I have NEVER said this is impossible. However, I will CONTINUE to say it's WAY more complicated than most of you think. THAT IS THE REASON IT HASN'T BEEN DONE.
The only thing so far I see you guys doing differently is enabling big maps and inserting some of tephra's code into the ROM for additional functionality. Rescaling maps has already been tried and has been PROVEN not to work. There is underlying code in the ROM that MUST be edited in order for a N/A ECU to understand boost. Otherwise, no matter what you do, you can only generate up to 100% load.
Do I fully understand the underpinnings of the ROM? No. Do I know from empirical evidence that the N/A ECU is not able to do any fueling or timing adjustments for load cells above 100%? Yes.
Ceddy: I am not confused and I don't work on Hondas. I PROBABLY am using improper terms for certain things, but don't take that for confusion.
Explain more about the correlation between load as listed in the table vs how the actual ECU reads airflow. What I've been trying to get across is that the evidence we have at hand from almost 5 years of using ECUFlash w/ the 3G computers is that SOMETHING IN THE CODE WILL NOT ALLOW ANY PORTION OF A TABLE TO BE READ OVER 100% LOAD. Why? Because the computer NEVER THINKS IT'S GETTING THERE.
If I rescale a map so that 1/2 of it is for load > 100%, I have effectively cut the map resolution in 1/2 because the 1/2 that is for load >100% will never be used. This has been verified countless times on numerous setups. Doesn't matter if it's the V6 ECU or the I4. Doesn't matter if it's a Galant, a Diamante, an Eclipse or even a Stratus Coupe... All the same thing!
Common sense dictates that if someone who has dedicated a good portion of his life to a single manufacturer, let alone a single platform tells you something, you would listen.
So I will say it one last time. In order for a N/A ROM to understand boost, THE ENTIRE CODE WILL NEED TO BE ANALYZED TO FIND OUT WHAT IS PREVENTING IT. It could be something as simple as a g/rev limitation. I do not know what causes it but it is real and MUST BE DEALT WITH or this project will not meet with success.
After all, I haven't been playing w/ a tactrix cable and 3G ECU's or ROMs since they first came out. I haven't tried rescaling maps beyond 100% load to see if I could actually get the ECU to read higher into the table.
I have NEVER said this is impossible. However, I will CONTINUE to say it's WAY more complicated than most of you think. THAT IS THE REASON IT HASN'T BEEN DONE.
The only thing so far I see you guys doing differently is enabling big maps and inserting some of tephra's code into the ROM for additional functionality. Rescaling maps has already been tried and has been PROVEN not to work. There is underlying code in the ROM that MUST be edited in order for a N/A ECU to understand boost. Otherwise, no matter what you do, you can only generate up to 100% load.
Do I fully understand the underpinnings of the ROM? No. Do I know from empirical evidence that the N/A ECU is not able to do any fueling or timing adjustments for load cells above 100%? Yes.
Ceddy: I am not confused and I don't work on Hondas. I PROBABLY am using improper terms for certain things, but don't take that for confusion.
Explain more about the correlation between load as listed in the table vs how the actual ECU reads airflow. What I've been trying to get across is that the evidence we have at hand from almost 5 years of using ECUFlash w/ the 3G computers is that SOMETHING IN THE CODE WILL NOT ALLOW ANY PORTION OF A TABLE TO BE READ OVER 100% LOAD. Why? Because the computer NEVER THINKS IT'S GETTING THERE.
If I rescale a map so that 1/2 of it is for load > 100%, I have effectively cut the map resolution in 1/2 because the 1/2 that is for load >100% will never be used. This has been verified countless times on numerous setups. Doesn't matter if it's the V6 ECU or the I4. Doesn't matter if it's a Galant, a Diamante, an Eclipse or even a Stratus Coupe... All the same thing!
Common sense dictates that if someone who has dedicated a good portion of his life to a single manufacturer, let alone a single platform tells you something, you would listen.
So I will say it one last time. In order for a N/A ROM to understand boost, THE ENTIRE CODE WILL NEED TO BE ANALYZED TO FIND OUT WHAT IS PREVENTING IT. It could be something as simple as a g/rev limitation. I do not know what causes it but it is real and MUST BE DEALT WITH or this project will not meet with success.
Mar 1, 2010, 08:44 AM
#19
Newbie
Join Date: Aug 2008
Location: IL
Posts: 25
Likes: 0
Received 0 Likes
on
0 Posts
I started looking into disassembly for my own personal use on a 2000 eclipse. I had no prior background in programming, I was able to come a long way by reading things like the h8/539 manual, the unofficial guide to ida pro, and most importantly evoM. I still am light years behind you guys. I have kind of lost interest in the car, but I would like to see this go forward, because there are people on club3g like the original poster that could benefit from further diaassembly and developement.
However, there are people out there that do not share my sentiment. They would like to goat evoM into sharing information, would like to use any advancement for personal gain, discourage open sharing of information, act like people that do not know are stupid newbs, and show how big their e-penor is while doing it.
Anyway, I think symptoms pharmecis are describing are similar to the IPW and load limits as seen on the evo. Also, I think Ceddy has removed them on the 2g flashable ecus.
However, there are people out there that do not share my sentiment. They would like to goat evoM into sharing information, would like to use any advancement for personal gain, discourage open sharing of information, act like people that do not know are stupid newbs, and show how big their e-penor is while doing it.
Anyway, I think symptoms pharmecis are describing are similar to the IPW and load limits as seen on the evo. Also, I think Ceddy has removed them on the 2g flashable ecus.
Mar 1, 2010, 05:54 PM
#20
Newbie
Thread Starter
Join Date: Apr 2009
Location: Arvada, CO
Posts: 14
Likes: 0
Received 0 Likes
on
0 Posts
However, there are people out there that do not share my sentiment. They would like to goat evoM into sharing information, would like to use any advancement for personal gain, discourage open sharing of information, act like people that do not know are stupid newbs, and show how big their e-penor is while doing it.
Once it is finished it will be posted on a few different communities with direct linking to this thread as well as a donation link for the developer who has put in a lot of time on this...no financial gain on my behalf from this endeavor.
We have however at this point found the Evo IX and 05 Eclipse ROMs share the identical load calculations though, and with 2byte load it is seeing loads in excess of 100% (140 to be specific). There is more to be done, but for the most part this is moving forward to the point of becoming a completely functional flash.
Thank you again to everybody who has given input and time for this so far, I only hope to someday be "up to speed" to give back at some point for others as well.
Mar 1, 2010, 10:57 PM
#21
Evolved Member
SPD_FRK and other VR4 people might like to explore the Tuning Guide I've done for the VR4 ans LEGNUM using EcuFlash, see the link below:
http://ozvr4.com/forums/showthread.php?t=6650
http://ozvr4.com/forums/showthread.php?t=6650
Thread
Thread Starter
Forum
Replies
Last Post
(GSX95)
Mitsubishi Vehicles (Airtrek / Outlander)
0
Nov 26, 2012 06:51 PM
