Tuning Evo789 ACD
Jan 8, 2014, 02:58 PM
#31
Evolved Member
Thread Starter
Floppyz, your Evo8 401403 xml is not up to date, I can see it still has the 4G scaling and I dont know what else.
You had me very confused there until I realised the pix showed two different roms.
TUNE NOTES is a segment of rom at the end where you can type in some tuning notes for the rom variant. eg what the 3 modes are designed for, or if for circuit or gravel rally or whatever.
It is unfortunate that the Ascii scaling spreads the display out, making it difficult to read.
I generally just enter data in this field using my hex editor XVI32.
An xml for 401403 has been added to the first post.
You had me very confused there until I realised the pix showed two different roms.
TUNE NOTES is a segment of rom at the end where you can type in some tuning notes for the rom variant. eg what the 3 modes are designed for, or if for circuit or gravel rally or whatever.
It is unfortunate that the Ascii scaling spreads the display out, making it difficult to read.
I generally just enter data in this field using my hex editor XVI32.
An xml for 401403 has been added to the first post.
Last edited by merlin.oz; Jan 8, 2014 at 03:07 PM.
Jan 9, 2014, 07:19 PM
#36
theres the differences, I cant comment on the grand sum of changes as Im still sifting. But you can easily download and compare the roms and read the axis to deicpher. Some things have been adjusted and to me it looks more than mild, but I dont exactly know how much weight some of these values have myself either..
Im gonna flash this at some point.
Jan 10, 2014, 03:07 AM
#39
Evolving Member
theres the differences, I cant comment on the grand sum of changes as Im still sifting. But you can easily download and compare the roms and read the axis to deicpher. Some things have been adjusted and to me it looks more than mild, but I dont exactly know how much weight some of these values have myself either..
Im gonna flash this at some point.
Jan 10, 2014, 11:12 AM
#41
Floppyz, your Evo8 401403 xml is not up to date, I can see it still has the 4G scaling and I dont know what else.
You had me very confused there until I realised the pix showed two different roms.
TUNE NOTES is a segment of rom at the end where you can type in some tuning notes for the rom variant. eg what the 3 modes are designed for, or if for circuit or gravel rally or whatever.
It is unfortunate that the Ascii scaling spreads the display out, making it difficult to read.
I generally just enter data in this field using my hex editor XVI32.
An xml for 401403 has been added to the first post.
You had me very confused there until I realised the pix showed two different roms.
TUNE NOTES is a segment of rom at the end where you can type in some tuning notes for the rom variant. eg what the 3 modes are designed for, or if for circuit or gravel rally or whatever.
It is unfortunate that the Ascii scaling spreads the display out, making it difficult to read.
I generally just enter data in this field using my hex editor XVI32.
An xml for 401403 has been added to the first post.
What I mean is that befor last youre update every roms (evo 7 8 9) has almost 100% values in Lockup mar vs thottle, and almost 0% in reduce lock vs G (logically increasing as the laterl G force increese as well).
Now that values are swapped. Do you mean that the reduceing of the lockup clutch ar always at 100% in avery condition of later G force?? how is it possibile?
sorry for my bad english, I hope that you can understand my doubt!
bye
Jan 10, 2014, 12:37 PM
#42
Evolved Member
Thread Starter
Floppyz, not sure i follow you, but, typically you wont or should not ever see an A-Map (Throttle input) with 100%. They are generally 0 on the left side of the map and progressing to something like 10% to 30% at full throttle.
There are some few exceptions, gravel rally tunes often have some high values from 40% to 100% throttle at low speed. That is so they can get lockup to drive back onto the road after a spin and or for launch.
B-Maps (G-Force) typically reduce lockup as lateral G increases.
Does that help?
So, if you see something way different to the above, its likely the xml is wrongly defined/labeled/addressed. There have been xml errors, lots of them.
I know the first set I sent to Colby years ago have many of errors. Its a learning process ans I think we are still a long way from knowing it all.
There are some few exceptions, gravel rally tunes often have some high values from 40% to 100% throttle at low speed. That is so they can get lockup to drive back onto the road after a spin and or for launch.
B-Maps (G-Force) typically reduce lockup as lateral G increases.
Does that help?
So, if you see something way different to the above, its likely the xml is wrongly defined/labeled/addressed. There have been xml errors, lots of them.
I know the first set I sent to Colby years ago have many of errors. Its a learning process ans I think we are still a long way from knowing it all.
Last edited by merlin.oz; Jan 10, 2014 at 12:42 PM.
Jan 10, 2014, 01:09 PM
#43
Evolved Member
Thread Starter
A bit of background on these latest xmls and the scaling changes for G-Force and Steering Angle.
I recently got hold of a Ralliart Evo7 Technical Manual, there were a few gems in there which clarified the G range and the steering angle detection.
The G sensors are a 0-5 volt output linear device (as you would expect).
2.5 volts is the 0 G point, with a swing of +/- 1.5 G.
1.0 v = 1.5G Acceleration
1.5 v = 1.0G Acceleration
2.5 v = 0 G
3.5 v = 1.0G Deceleration
4.0 v = 1.5G Deceleration
When I normalized the G-Force axis to 1.5G it all came out perfect with sensible values.
The very next page in the manual showed the steering sensor, with a minimum angle detection (slit width) of 4*, with the following 4* ocolulted. So each pair equates to 8*.
360* / 8* = 45. The raw max value in the steering axis is 450, so the scaling I believe is x*0.8. The steering wheel will actually go a bit past 360* but I dont think they were worrying about that for the A/B/C maps.
The AYC map has a different steering angle axis, which goes out to decimal 512, which with the same scaling is 410*, which seems about right for steering wheel maximum rotation.
AYC control preceded ACD (Evo6 etc) and Mitsu decided to keep the same AYC code and table/axis arrangement for the Evo789. Thats my guess anyway.
I recently got hold of a Ralliart Evo7 Technical Manual, there were a few gems in there which clarified the G range and the steering angle detection.
The G sensors are a 0-5 volt output linear device (as you would expect).
2.5 volts is the 0 G point, with a swing of +/- 1.5 G.
1.0 v = 1.5G Acceleration
1.5 v = 1.0G Acceleration
2.5 v = 0 G
3.5 v = 1.0G Deceleration
4.0 v = 1.5G Deceleration
When I normalized the G-Force axis to 1.5G it all came out perfect with sensible values.
The very next page in the manual showed the steering sensor, with a minimum angle detection (slit width) of 4*, with the following 4* ocolulted. So each pair equates to 8*.
360* / 8* = 45. The raw max value in the steering axis is 450, so the scaling I believe is x*0.8. The steering wheel will actually go a bit past 360* but I dont think they were worrying about that for the A/B/C maps.
The AYC map has a different steering angle axis, which goes out to decimal 512, which with the same scaling is 410*, which seems about right for steering wheel maximum rotation.
AYC control preceded ACD (Evo6 etc) and Mitsu decided to keep the same AYC code and table/axis arrangement for the Evo789. Thats my guess anyway.
Jan 10, 2014, 01:53 PM
#44
Ok Marlin, that what I espected.
So the xml definition maps we worked on till 2 days ago where wrong... Cose values in map A where in Maps B...
I've read your notes in xml files; this is what I think as ex (happy) rally driver:
- A maps (throttle vs lock) as you say should increese as you push more on throttle; that value should decrease as the speed increese, in higher gear there is less torque on weels so less slippage. In my opinion in low grip surface (gravel and snow) the lockage factor must decreese, less torque required for slippage. So less lockage on WOT but a little bit early start lock on middle throttle. On high grip you need more power for lock the spinning, but the spinnig start later
- B maps (G vs lock reduce) as you've said should decreese lockage power as the G factor increese. This value, of reduceing lockage power, should increese in low grip surface (gravel snow): less grip, less lateral G less torque on weels.
- C maps (lockup vs steer angle) this map should reduce the lockage power on high steering angle, that for reduce the understeering. Same as before: low grip, less lockage need. This map shoud be intresting for differetial entering in the corner (decel moment) where the understeering is to fight 'couse you must to turn the car (better a little oversteering while brakeing inside the corner), and the exit from the corner (accelereting) where the oversteering makes you loose speed (better a little bit of understeering).
Any ideas for others 2d maps? (D, E,F ... etc)
bye
So the xml definition maps we worked on till 2 days ago where wrong... Cose values in map A where in Maps B...
I've read your notes in xml files; this is what I think as ex (happy) rally driver:
- A maps (throttle vs lock) as you say should increese as you push more on throttle; that value should decrease as the speed increese, in higher gear there is less torque on weels so less slippage. In my opinion in low grip surface (gravel and snow) the lockage factor must decreese, less torque required for slippage. So less lockage on WOT but a little bit early start lock on middle throttle. On high grip you need more power for lock the spinning, but the spinnig start later
- B maps (G vs lock reduce) as you've said should decreese lockage power as the G factor increese. This value, of reduceing lockage power, should increese in low grip surface (gravel snow): less grip, less lateral G less torque on weels.
- C maps (lockup vs steer angle) this map should reduce the lockage power on high steering angle, that for reduce the understeering. Same as before: low grip, less lockage need. This map shoud be intresting for differetial entering in the corner (decel moment) where the understeering is to fight 'couse you must to turn the car (better a little oversteering while brakeing inside the corner), and the exit from the corner (accelereting) where the oversteering makes you loose speed (better a little bit of understeering).
Any ideas for others 2d maps? (D, E,F ... etc)
bye
Last edited by Floppyz; Jan 10, 2014 at 02:15 PM. Reason: ...
