Open source Hybrid Speed Density

[Yobiwan]

thanks a lot , @ztan.

I apply this to my car today. and it works great . (Korea 6MT A01E rom )

Left graph is Air g/s from MAP sensor by ztan's SD method,

and Right one is Air g/s from existing maf sensor.

less variance , smoother line.

I will install forged short block next month, and crank up boost slightly,

and this SD will help me alot.

[Yobiwan]

I think no one cares about SD tuning with OFT platform now .

but it works flawlessly !

this is my final SD table & log of WOT 3-4-5 gear.

stock motor
SBD500x
DW450 injector


https://datazap.me/u/yobiwan/sbd-v58...zoom=2239-2474

[tomm.brz]

That's great Yobiwan,
is it a pure SD so? or hybrid?

I hate SD on ecutek, the volumetric efficiency table is annoying to tune, and that SD MAF table you have would be a lot better


a little off topic, i see you keep -0,5 of minimum pull timing for flkc and knock correction
Do you think is safe enough?
i tend to keep -0,7 minimum for NA, and -1,05 for FI

[Yobiwan]

Originally Posted by tomm.brz

That's great Yobiwan,
is it a pure SD so? or hybrid?

I hate SD on ecutek, the volumetric efficiency table is annoying to tune, and that SD MAF table you have would be a lot better


a little off topic, i see you keep -0,5 of minimum pull timing for flkc and knock correction
Do you think is safe enough?
i tend to keep -0,7 minimum for NA, and -1,05 for FI

yes , it's hybrid SD that ztan made above.

under 120g/s 100% MAF
120~200g/s MAF + SD blending
above 200g/s 100% SD

this SDMAF table is very easy to tune. almost same as MAF scale.
boost , AFR , Command AFR WOT log are all that needed for scale.
We can ignore under 120g/s area because MAF will cover that area.

It's 100% street car. and I set fine correction retard value at -0.5 after turbo installation.
so -0.5 flkc is minimum value.
I pull timing when any flkc is on 3rd gear WOT or -1 FLKC on any gear WOT.

[Kodename47]

Originally Posted by tomm.brz

I hate SD on ecutek, the volumetric efficiency table is annoying to tune, and that SD MAF table you have would be a lot better

Really?? I think the EcuTek VE table is rather simple as it works in VE % rather than a MAF translation. The calculations you need to tune it are no different to those you would need for the table above.

It's more like the sort of tables you would find on an aftermarket ECU, so most tuners are used to those....

[tomm.brz]

you know what, kodename... i tune using a lot your maf scaling tool ) now i'm supposed to thank you, and I do
and i haven t found a way yet to import the SD VE map to your tool from ecutek
So when I use SD i manually adjust the VE table and the result is always kinda sloppy specially under 0.9 of manifold pressure (for NA, in this example) and it takes so much time

an SD MAF like ztan's one just seems more familiar to me sorry

[steve99]

Originally Posted by tomm.brz

That's great Yobiwan,
is it a pure SD so? or hybrid?

I hate SD on ecutek, the volumetric efficiency table is annoying to tune, and that SD MAF table you have would be a lot better


a little off topic, i see you keep -0,5 of minimum pull timing for flkc and knock correction
Do you think is safe enough?
i tend to keep -0,7 minimum for NA, and -1,05 for FI

With ecutek SD try enabling full time closed loop fueling, then observe the closed loop fuel correction in the area if its positive add that percent to the SD VE table if it negitive subtract. Ie use the cl correction like you would with ltft correcting maf scale


Assuming your injector scaling is correct and your 02 sensor is rescaled to read rich enough

[tomm.brz]

i already do this way, actually if you enable CL fuel correction all the time even at low engine loads, you would need to zero Af#1 correction or the STFT works against it and it become an impossible mess... anyway it s just time consuming the correction of the sd ve map
Any easy method to import SD VE table from ecutek to the maf scaling tool of vgi and kodename?
when i paste it it just gives error

[steve99]

Originally Posted by tomm.brz

i already do this way, actually if you enable CL fuel correction all the time even at low engine loads, you would need to zero Af#1 correction or the STFT works against it and it become an impossible mess... anyway it s just time consuming the correction of the sd ve map
Any easy method to import SD VE table from ecutek to the maf scaling tool of vgi and kodename?
when i paste it it just gives error

Yes when. Suabaru closed loop and ecutek closed loop on togeather they tend to fight with each other, usually set ecutek closed loop like @Tor to bot work till 0.7 load

[tomm.brz]

i prefer 0.9 with hysteresis of 0.1, with 0.7 i had moments in closed loop with stft working and CL ecutek fighting against it

[Kodename47]

I don't know what is causing the error on the SD tab, I haven't tried it for a while - have you downloaded a more up-to-date version as I know Vgi occasionally updates it for bu fixed?

I do all these calculations manually by adding in a coupe of manually calculated columns (dV/dt and error %) and then use the Log Stats tab to give me the error outputs and a spreadsheet to make the changes. That way I have more control over some of these outputs using the filters than the default tabs allow.

[1quickchic]

Originally Posted by Yobiwan

I think no one cares about SD tuning with OFT platform now .

Actually, I am, but I have yet to completely figure out how to implement it, I don't suppose you'd be willing to assist me?

Right now I am being held back by the limitations of not having sd, so

FYI my rom is a01d us auto

[Yobiwan]

Originally Posted by 1quickchic

Actually, I am, but I have yet to completely figure out how to implement it, I don't suppose you'd be willing to assist me?

Right now I am being held back by the limitations of not having sd, so

FYI my rom is a01d us auto

Sorry I can’t help that.
Only rom that we can use SD on OFT platform is 6MT (2012-2017) by now.

[1quickchic]

@Yobiwan ,well here is a question, is it possible for me to make the changes to my ecu?a way to either import the settings or to write them into the rom, I also can use tactrix if its a better solution