Map editing

Discussion of tuning specific to MAZDASPEED3/MAZDASPEED Axela/Mazda 3 MPS vehicles
mituc
VersaTuner guru
Posts: 1324
Joined: December 17th, 2011, 2:47 pm
Location: Iasi/Romania

Re: Map editing

Post by mituc »

EmPeEs6 wrote: July 3rd, 2020, 11:25 pm I know, but is it used (directly) before throttle closing operations or simultaneously?
What exactly is the corresponding "closed loop" mechanism to limit boost under overboost conditions - this is still totally unclear...
This mechanism works regardless of the ECU operation mode, CL or OL.
EmPeEs6 wrote: July 3rd, 2020, 11:25 pm The principle meaning is clear to me, but if they are real limiting tables, I would assume that the ecu not just refers to the req. load x gear tables as final target load (if no other target/limiter table value is set lower) but also tries to keep load below this value under all circumstances, e.g. using throttle position adjustments.
Think about the load target as the hole in a funnel. If the water you try to pour in goes up, down, left, right, whatever, the walls of the funnel will still direct the water flow through the hole. So that's how load target works.

Of course the real target depends on various other tables so the actual target is the lowest value of all.
EmPeEs6 wrote: July 3rd, 2020, 11:25 pm But if the ecu only allows/accepts a max value of 20, why is it possible to enter values up to 100?
This value in particular is not about what the ECU allows or accepts, but about what's practical and about how the learning/adaptation vectors are being used. It's like targeting load 4 on a stock turbo: the ECU will try to reach it assuming other limits are not met, but because that's not actually possible you may end up hitting load 2 at 1/3 of the APP (which will eventually correspond to 78% throttle position).
EmPeEs6 wrote: July 3rd, 2020, 11:25 pm Unfortunately, this is not true if there are unknown/unavailable tables, which may have a significant influence on the load targeting logic or any inconsistencies or even "bugs" in the ecu programming logic (e.g. the fact that my ecu simply ignores the target load values for app 100).
What is the table you use to target load based on APP? Or you're actually thinking about the whole process, APP tables -> drive by wire tables -> load?
2008 Cosmic Blue Mazda 3MPS
Built engine + WMI + GTX3071 gen2, ~509BHP @34PSI
2008 Icy Blue Mazda CX7
Stock engine and exhaust (YES!!), JBR3" + GTX2867 gen2 + Autotech HPFP, self-tuned to 330-ish BHP
EmPeEs6
Posts: 20
Joined: November 25th, 2013, 5:54 pm

Re: Map editing

Post by EmPeEs6 »

The still greatest riddle to me is the concrete ECU usage of the "BCS DC Adjustments - Boost Limit Exceeded" table - all other tables, which are relevant for achieving the target loads are quite clear (even if I doubt that anyone on this earth really únderstands all accessible tables in detail).
Meanwhile, I think that this table is not just used during pure overboost conditions, but more or less constantly when the ECU is trying to limit boost (and thus load) using throttle closing operations (in addition to wgdc adjustments) in order to reach a certain target or stay below some limit!
This would make sense since the values in this table correspond (concerning their amount) to the boost and load error correction values which are aplied a few hundred times per rpm in order to reach the desired (calculated) target load.
mituc
VersaTuner guru
Posts: 1324
Joined: December 17th, 2011, 2:47 pm
Location: Iasi/Romania

Re: Map editing

Post by mituc »

EmPeEs6 wrote: July 4th, 2020, 9:48 pm The still greatest riddle to me is the concrete ECU usage of the "BCS DC Adjustments - Boost Limit Exceeded" table - all other tables, which are relevant for achieving the target loads are quite clear (even if I doubt that anyone on this earth really únderstands all accessible tables in detail).
I think I already mentioned this: when the boost reaches or goes above the values in the Absolute Limits -> Overboost Protection Throttle Reduction.
At the same time when throttle reduction is applied you will see WGDC corrections as well.
I tested this in countless situations and if you want you can do a test yourself: set the values in the overboost protection throttle reduction to a ridiculously low value, say 10PSI, and go mid/high POT.
EmPeEs6 wrote: July 4th, 2020, 9:48 pm Meanwhile, I think that this table is not just used during pure overboost conditions, but more or less constantly when the ECU is trying to limit boost (and thus load) using throttle closing operations (in addition to wgdc adjustments) in order to reach a certain target or stay below some limit!
It actually is only for overboost mitigation. If none of the boost limits was exceeded that table will not be used for any correction. For anything else you have the Load target/limit tables, APP requested load tables and the Drive by wire tables. If somewhere in these tables you request a certain throttle opening for a certain load value and that value is too high or too low you will see the same BCS DC load correction/compensation tables but NOT the overboost protection tables.
EmPeEs6 wrote: July 4th, 2020, 9:48 pm This would make sense since the values in this table correspond (concerning their amount) to the boost and load error correction values which are aplied a few hundred times per rpm in order to reach the desired (calculated) target load.
Well, the correction values are not applied a few hundred times per RPM, but more rare than this. The sensors are not that fast, some actuators (think wastegate, throttle) and slow reacting. If the comp/correction tables have values that are too aggressive it may lead to heavy oscillations around the target, if they are too low then you may hit a cut because the boost/load did not stop when throttle and wastegate corrections were applied.
By example during a WOT pull you may start overboosting at 4500rpm and you will see a but or corrections 500-700rpm later, especially in lower gears (2/3).
Some other times you may see in your logs that the fuel cut overboost protection limits were exceeded 3-4-5 log lines consecutively and you barely felt a mild tremor instead of a violent faceplant-to-the-steeringwheel fuel cut. These are slow reacting systems. getting the values right from the get go goes a long way.
2008 Cosmic Blue Mazda 3MPS
Built engine + WMI + GTX3071 gen2, ~509BHP @34PSI
2008 Icy Blue Mazda CX7
Stock engine and exhaust (YES!!), JBR3" + GTX2867 gen2 + Autotech HPFP, self-tuned to 330-ish BHP
EmPeEs6
Posts: 20
Joined: November 25th, 2013, 5:54 pm

Re: Map editing

Post by EmPeEs6 »

This is why the correction values should not be symmetric (to prevent endless oscialltions around your target load), not too low (in order to enable the system to reach the target load in time) and not too high (to avoid overboosting).
The best setting would be very little adjustments within a small correction range (which is unfortunately not possible using the stock ebcs when you have a wide wgdc range during wot runs from e.g. 40 to almost 100).
mituc
VersaTuner guru
Posts: 1324
Joined: December 17th, 2011, 2:47 pm
Location: Iasi/Romania

Re: Map editing

Post by mituc »

EmPeEs6 wrote: July 5th, 2020, 1:45 am This is why the correction values should not be symmetric (to prevent endless oscialltions around your target load), not too low (in order to enable the system to reach the target load in time) and not too high (to avoid overboosting).
The simmetry has nothing to do with this. The situation you are trying to mention is when the corrections are too aggressive and you jump from a possible correction to the symmetric negative one. This is a clear case for overshooting those correction values in the table. If you are in this situation then you need to lower the values in the table that is being used for corrections in your case.
EmPeEs6 wrote: July 5th, 2020, 1:45 am The best setting would be very little adjustments within a small correction range (which is unfortunately not possible using the stock ebcs when you have a wide wgdc range during wot runs from e.g. 40 to almost 100).
Wait, if you are using the stock EBCS then the best choice is to use the stock comp tables. Things change quite a bit when using a 3 port MAC valve based EBCS (grimpseed, cobb, CS, Stratified, whatever) because that is a lot more reactive than the factory one and often when paired with a slower responding turbo things get more complicated in controlling load/boost.
2008 Cosmic Blue Mazda 3MPS
Built engine + WMI + GTX3071 gen2, ~509BHP @34PSI
2008 Icy Blue Mazda CX7
Stock engine and exhaust (YES!!), JBR3" + GTX2867 gen2 + Autotech HPFP, self-tuned to 330-ish BHP
CarbonMica
Posts: 1
Joined: July 6th, 2020, 11:48 am

Re: Map editing

Post by CarbonMica »

Hello to All. 2007 bought not running, now fine. Stock Turbo, Tubing (unknown) Exhaust Header, Stock Turbo, Cobb Big Mouth 2.87"CAI, Cobb Intake Tube Type Manifold, MAF cal done, 3 Bar MAP, Grimmspeed BCS/ Interrupt. HKS HOV plumbed into Cobb Intake
Currently using nicely, pulls and logs look good. Fine Tuning needed---- I am asking questions----
Base93US/93EU Octane -Grimmspeed BCS Interrupt Tune.
I want to try some other Tunes especially when I install a cheap 3" Downpipe. Yes at 120$ was a steal and my time is free.
I will replace Turbo at some point, on the cheap.
What do I need to Modify on other tunes in Easy Tune Database to use Grimmspeed BCS plumbed Interrupt. ? Should I just remove...
I'm cheap :facepalm if I have to remove to try other Tunes?
No one area in Tune editor for easy modification of Easy Tune Database for Grimmspeed BCS option that I can find. So mod's of existing Easy Tunes looks best way.

Thank You for help or Knowledge to say NO!!!

Edit 1> Can I just use Installed catless downpipe" option when flashing the tune? Found...yes looked at other posts :smashpc so thought to add if this all I need to do for added Downpipe ?
Post Reply