The first step of reflashing a 32bit ECU is to erase the stock map/software image. There is no provision for backing up the stock image or unerasing it. Make sure you flash the correct file and know what you are doing before using these files.
"Petrik has a more detailed definition but he has stated the he will no longer contribute updates to the site so I consider it an unsupported version that may have errors."
Yep, right now the personal focus is in building my own bike. Meanwhile I am continuing to write the simple ecueditor software for tuning the bike. I consider that most of the variables and maps are now known and hence no regular updates are right now not happening to the .xml.
__________________
When asking a question, you can also consider posting it to facebook:
I've updated the definition so it now includes all the maps listed in map table. There were 87 2D maps and 53 3D maps for a total of 140.
I labled all the large 3d maps, ignition, rpm v TPS (alpha-n), rpm v MAP (speed density) fuel etc.
I identified the ignition maps by gear, bank and map select. PetriK had already done much of this in his preliminary definition but I wanted to go back through it myself to make sure I understand what is going on. Unlike some of the other bikes that have seperate 3D rpm vs TPS maps for neutral this ECU actually use a 2D map for neutral that is advance by RPM. It ignores the TPS when in neutral.
Also I found a map switch to turn off the clutch pulled in ignition feature. On the stock bike when you pull in the clutch it uses the 2d Neutral map. When you disable this with the map switch the map is selected the same whether the clutch is in or out, by gear.
I also found a map switch that makes 2nd gear use the 1st gear maps. In the stock bike this is disabled so the ignition maps are: 1st, 2nd, 3&4, 5&6. With it enabled it becomes 1&2, 3&4, 5&6.
On the difference between MS0 and MS1 map sets I used the same convention I always use. When PortH, bit4=0 the maps selected are MS0. In Petrik's early definition he labled the MS0 maps MSon and the MS1 maps MSoff. I don't know if we have a differrent interpetation of on and off or if he found the signal was inverted between the wire harness and pin and PortH when he traced it out.
In Petrik's early definition he labled the MS0 maps MSon and the MS1 maps MSoff. I don't know if we have a differrent interpetation of on and off or if he found the signal was inverted between the wire harness and pin and PortH when he traced it out.
That is one of the peculiarities I have noticed. When running the EU bin on US firmware the MS signal become inverted. I have implemented the MS maps tuning on ECUeditor and the definition I used there works for EU bin and both EU hw. It is possible that due to dremeling the pcb I may have something wrong with my US hw board, but anyhow because of small uncertain items like this I consider ECUeditor not being fully tested for US bins.
__________________
When asking a question, you can also consider posting it to facebook:
SO NOW MS0 = PIN 21 GROUNDED, MS1 = PIN 21 OPEN, RIGHT?
Correct. It takes a really convoluted path to get there but in the end it is the same as every other Denso ECU I've looked at
The pin grounded = low = 0 = MS0 (the first mapset with the lower numerical address) The pin open = high = 1 = MS1 (the second mapset with the higher numerical address)
I just checked cranking limits to improve starting of big cc & higher compression engines. That made me to check the RPM calculation algorithm and verify it by debugging. Unless I understood something incorreclty, the RPM is calculated in sub Calculate_RPM_FFFF8398_sub_80DE:
In that it looks like the 4 * RPM = 4 * 3840000/Revolution time. The formula we have assumed for calculating RPM ((37500/x)*2048)/5.12 = 15000000 where as 4*384000 = 1536000.
There may be something I have lost in bitwise division somewhere (1536000/1.024=15000000) or is it possible that we have 2.4% error in some of our RPM numbers ?
__________________
When asking a question, you can also consider posting it to facebook:
Ram location 0x83B8 or what I call CrankPeriod holds the number of timer ticks that equal 180 degrees of rotation. If the timer is running at 2uS per tick and the CrankPeriod count is 1389 then the 360 degree rotation period is...
1389 * 2 * 0.000002 = .005556 S
Divided by 1 sec = 179.9856012 / S (rotations per second) Times 60 = 10799.13607 or 10,800 rotations per minute.
When I plug the value 1389 into the calculate RPM routine at 0x80DE the value saved as RPM is 0x6bFD or 27645.
27645 / 2.56 = 10798.828125
Meanwhile, (37500/1398) * 2048 / 5.12 = 10799.136
I think these are all close enough to call 10,800.
I'm not sure how your getting 4 * RPM = 4 * 3840000/Revolution time
RR, you are most likely correct. When monitoring ram values I thought I was getting:
FFFF8398/256*10 = 4 * 3840000/FFFF83B8
4*3840000/1389 = 11058 vs 4*3840000/1389/1.024=11799
... still dont have HEW 7052 simdebug up and running for testing individual subs like 0x80DE, so the division most likely happens with 3840000/1.024/Revolutiontime ?
__________________
When asking a question, you can also consider posting it to facebook:
Every Denso map I've ever seen the maps are listed with lowest numerical address being cylinder 1. In other words they are most likely already listed in cylinder order 1,2,3, 4
As to the second question the x13y8 maps are RPM - Throttle while the x15y8 are most likely RPM - Manifold Absolute Pressure.
The Manifold axis is the only ones I've ever seen where the numbers start high and end low left to right. I didn't actually check the code on the x15y8 but I'd bet money that's what it is.
Added to the misc items two switches that will disable the factory fuel trim values set using the 'Yoshbox' by setting the values to 0x80 which is +/- 0 %
Both switches must be set to the same state for it to work.
I haven't tested it on a bike but it sets the file bytes the same as the similar function in ECUeditor.
Im putting a Hayabusa motor in a small sportscar and would like to change revlimits and possibly do a slight retune since my installation differs from stock in regards of intake airbox and exhaust.
From what I understood its hard/impossible to read out the stock map from the ECU. The stock map available is refered to as an US map. Is an EU version available or are they the same? I assume that different emission legislations means different maps, but are there any performance differences between US/EU versions? Is it a bad idea to use a US map as a base for a slight retune of a EU-motor? (I assume that for a complete retune of all loadpoints, it doesnt matter what base map you use.)
Updated the Ram Air Compensation maps and added a Map Switch to disable the Ram Air Comp with a single click so you don't have to zero all six tables for a dyno tune then return them to stock when your done.
The function in the software that retrieves the ram air comp map value selects a map by gear position but has a default value if the bike is in neutral. The map switch I added forces the function to always return that neutral value.
Note that ECUeditor returns a Ram Air Comp value of 20 for all gears/rpms when in disabled mode. The RomRaider Ram Air Comp disable returns a Ram Air Comp value of 0 for all gears/rpms.
Housekeeping Note: I changed the name of the defintion from busa2002-2006.xml to GSX1300R_K2-7.xml. This definition works for 2002-2007 so I updated the file name. Make sure you remove the old version and add the new version under ECU definitions manager in RomRaider.
I´m just about to finetune my idle IAP maps. As far as I can see rpm of those maps go until 2000. When I look at the fuel map switching in RomRaider I can see that the upper rpm limit of the idle map is 2450. How is the amount of fuel calculated in that area (2000 - 2450 rpm)? Should I lower the idle map switching to 2000 rpm?
If I have been using the ecu editor and upgrade to this how much do I need to change? Or is this the same basic system using the same FTD?? Thank you for your help Ed
You would still use FTD to flash, ecu editor and rom raider are both ways of editing the .bin file, you still have to have external software to actually flash the ecu.