No direct reading of load from ECU?
Thread Starter
EvoM Guru
iTrader: (50)
Joined: Mar 2006
Posts: 9,675
Likes: 132
From: Tri-Cities, WA // Portland, OR
No direct reading of load from ECU?
Since everyone is calculating load indirectly, does that mean that the request ID for load has not yet been found, or the ECU does not provide load via any of the diagnostic protocols?
The ECU does calculate load.. Obviously.. However a standard requestID to specifically dump that value does not appear to exist.
The good news is there has been a pretty big effort to disassemble the rom, which means that at some point we will be able to do things with the ECU..
Plans for mods include:
Outputting a MAF Frequency that isn't clipped at 1609hz
Outputting the internal ECU's calculated load value
Additional resolution to the fuel and timing maps
Altering the OCTane value with an external device manually to allow dual map functionality (This obviously would remove a safety feature)
Some level of real time tuning when the RAM locations for the maps are decoded
Additionally there are 20 or so additional configuration items and tables that affect the engine management, some of which I have some clue what they are, others I have no idea..
Closed Loop AFR Target
Injector configuration parameters (Dead time, Latency, etc in addition to the Voltage latency adjustment)
One of the AktiveMatrix members in intent on creating some sort of speed density or alternative MAF sensor swap programming.
The 02-05 ECU has 256kb of rom data, functionally I think only a small part of it is used, and of course maps, tables, and configuration parameters are all spread around in there. At some point its feasable that enough will be known about the roms to be able to pick and choose the features you want and modify it in any way you like.
The good news is there has been a pretty big effort to disassemble the rom, which means that at some point we will be able to do things with the ECU..
Plans for mods include:
Outputting a MAF Frequency that isn't clipped at 1609hz
Outputting the internal ECU's calculated load value
Additional resolution to the fuel and timing maps
Altering the OCTane value with an external device manually to allow dual map functionality (This obviously would remove a safety feature)
Some level of real time tuning when the RAM locations for the maps are decoded
Additionally there are 20 or so additional configuration items and tables that affect the engine management, some of which I have some clue what they are, others I have no idea..
Closed Loop AFR Target
Injector configuration parameters (Dead time, Latency, etc in addition to the Voltage latency adjustment)
One of the AktiveMatrix members in intent on creating some sort of speed density or alternative MAF sensor swap programming.
The 02-05 ECU has 256kb of rom data, functionally I think only a small part of it is used, and of course maps, tables, and configuration parameters are all spread around in there. At some point its feasable that enough will be known about the roms to be able to pick and choose the features you want and modify it in any way you like.
Originally Posted by tabio42
Load Value; Unclamped MAF frequency or Speed Density Conversion FTW!
That would be the ultimate stock ECU.
That would be the ultimate stock ECU.
This was something obviously a throwback to the days with stock turbo cars rarely reading 1609hz at max.. Once your modified you frequently go over that, but back in the day of 16-18psi max, I think the highest you'd see would be around that.
I'm running a different kind of MAF sensor, its not a karmann sensor.. The honest truth is the karmann sensor is the most accurate MAF sensor available, its that they don't make one sized for the kind of airflow our cars are capable of with aftermarket turbos (and even really nicely tuned stock turbo variants)
Trending Topics
Originally Posted by FlashBlueRS
RequestID "1C" and eval "5x/8" should give you guys load. It works for the 3G eclipse.
I pulled that from the new ecuEdit release for Mitsu's.
I pulled that from the new ecuEdit release for Mitsu's.
Code:
<DataListItem DataLog="Y" Display="ECU Load raw" LogReference="ecuLoadraw" RequestID="1c" Eval="x" Unit="steps" /> <DataListItem DataLog="Y" Display="ECU Load" LogReference="ecuLoad" RequestID="1c" Eval="0.625x" Unit="%" />
Code:
ecuLoadraw ecuLoad Load 39 24.375 24.37516.993125 42 26.25 26.2516.993125 44 27.5 27.522.123125 46 28.75 28.7522.123125 139 82.5 82.522.123125 91 56.875 56.87563.163125 89 55.625 55.62547.773125 134 83.75 83.7558.033125 137 85 8578.553125 131 80.625 80.62578.553125 116 72.5 72.568.293125 55 33.75 33.7542.643125 33 20 2022.123125 27 16.875 16.87516.993125 29 18.75 18.75-8.458625954198473282442748092 38 23.75 23.75-8.394545454545454545454545455 41 25.625 25.62516.993125 38 24.375 24.37516.993125 39 24.375 24.37516.993125 40 25 2516.993125 42 26.25 26.2516.993125 43 26.875 26.87522.123125 45 28.125 28.12522.123125 45 28.125 28.12522.123125 45 28.125 28.12522.123125 45 28.125 28.12522.123125 45 28.125 28.12522.123125 46 28.125 28.12522.123125 46 28.75 28.7522.123125 47 28.75 28.7522.123125 46 28.75 28.7522.123125 144 96.875 96.87522.123125 57 33.125 33.125104.203125 78 46.875 46.87588.813125 139 88.125 88.12583.683125 144 89.375 89.37583.683125 141 88.125 88.12578.553125 141 87.5 87.578.553125 132 82.5 82.578.553125 144 90 9073.423125 159 99.375 99.37588.813125 172 107.5 107.593.943125 180 113.125 113.125100.05176470588235294117647059 193 121.25 121.25107.92987012987012987012987013 222 139.375 139.375118.41477707006369426751592357 255 159.375 159.375140.82904109589041095890410959 255 159.375 159.375166.60609271523178807947019868 255 159.375 159.375200.78430379746835443037974684 255 159.375 159.375221.46585365853658536585365854 255 159.375 159.375229.28335329341317365269461078 26 15 1553.542252 20 11.25 11.2516.354285714285714285714285714
I'm impressed.. I was always hoping there would be an actual load.. But I'd need to see a run where the load was around 220 or higher, outside the clipping range of the MAF/MDP sensor logging range to see if this works right..
Edit.. If you take the max value 255, you won't see a load over 159 if the formula is 0.625
and 159.375 if you use the 5x/8
Unfortunately this load value is also clipped, its impossible for it not to be..
However, it can be used as a check value to make sure the load calc algorithm is correct for your settings.. I guess this range might have applied if your max boost is under 14psi or so (typical for older DSM's)
Edit.. If you take the max value 255, you won't see a load over 159 if the formula is 0.625
and 159.375 if you use the 5x/8
Unfortunately this load value is also clipped, its impossible for it not to be..
However, it can be used as a check value to make sure the load calc algorithm is correct for your settings.. I guess this range might have applied if your max boost is under 14psi or so (typical for older DSM's)
Last edited by MalibuJack; Sep 24, 2006 at 06:39 PM.
Originally Posted by MalibuJack
I'm impressed.. I was always hoping there would be an actual load.. But I'd need to see a run where the load was around 220 or higher, outside the clipping range of the MAF/MDP sensor logging range to see if this works right..
Edit.. If you take the max value 255, you won't see a load over 159 if the formula is 0.625
and 159.375 if you use the 5x/8
Unfortunately this load value is also clipped, its impossible for it not to be..
However, it can be used as a check value to make sure the load calc algorithm is correct for your settings.. I guess this range might have applied if your max boost is under 14psi or so (typical for older DSM's)
Edit.. If you take the max value 255, you won't see a load over 159 if the formula is 0.625
and 159.375 if you use the 5x/8
Unfortunately this load value is also clipped, its impossible for it not to be..
However, it can be used as a check value to make sure the load calc algorithm is correct for your settings.. I guess this range might have applied if your max boost is under 14psi or so (typical for older DSM's)
But an older DSM (like my 90 eclipse) can use load values well into the 4grams/rev in the ecu AND send it to my pocketlogger. On the stock 1g ecu the load is a value from 0-255 decimal. What you do is divide this number by 128 to get a load value.
EDIT:
Who's to say that the x axis in ecuedit is labeled correctly? Why can't the load be from 0 to 3.00 grams/rev instead of a % value? Wheras the stock 1g DSM's were limited to 2.14 gram/rev (approximatley 215 wheel horsepower worth of "flow") it would make sence that an evo would have maps that go to 3.00 grams/rev (a value of 300% in ecuedit) or approximatley 300whp worth of "flow".
EDIT:
Who's to say that the x axis in ecuedit is labeled correctly? Why can't the load be from 0 to 3.00 grams/rev instead of a % value? Wheras the stock 1g DSM's were limited to 2.14 gram/rev (approximatley 215 wheel horsepower worth of "flow") it would make sence that an evo would have maps that go to 3.00 grams/rev (a value of 300% in ecuedit) or approximatley 300whp worth of "flow".
Last edited by dan l; Sep 24, 2006 at 06:56 PM.