• Hello Guest, welcome to the initial stages of our new platform!
    You can find some additional information about where we are in the process of migrating the board and setting up our new software here

    Thank you for being a part of our community!

960/S90 M4.4 XDF...work in progress.

Ok this is exciting. Haven't had the time to put together my B6304S+T swapped car yet, but I had been debating pulling the trigger on an MS3 for the swap since I have a complete 960 harness installed in the car. Definitely interested in any updates that come along and once the car is together can do some testing.
 
Ok this is exciting. Haven't had the time to put together my B6304S+T swapped car yet, but I had been debating pulling the trigger on an MS3 for the swap since I have a complete 960 harness installed in the car. Definitely interested in any updates that come along and once the car is together can do some testing.
It's possible that it could be quite simple. The stock MAF is obviously not overflowed at stock power, so there's a bit of wiggle room for some boost. Not a lot, but it's a good start. Adjusting the timing maps would make it safer to run boost as well. I've done these sort of mods on VW motronic 2.9 bin files before and despite being driven like it was a stolen car, it held up nicely :).
 
I feel like this is doable, considering guys in Europe turbo these things without even tuning the ECU. Seems its a very forgiving system. I am also excited about not spending $1000 on MS3 if I don't have to. :lol:
 
As it stands right now, the XDF that's been posted on the first page somewhere (can't remember where), is 'complete' enough to adjust for larger injectors, and POSSIBLY adjust the timing curve. That's something I'd highly recommend doing, if you're using the stock B6304S with it's 10.7:1 engine. Mine is holding up to 7 psi so far :).
 
M4.4 is almost equal to standalone management actually we can almost do anything with it, even run it on wideband control, mapswitching during driving, it even does launch control and cops on 5 cyl.

The great benefit of the system is that it's with adaptive knock control, so it's "self learning" on timing within the limitations of the system ofcourse.

But any cheap ms, speeduino, kdfi stuff don't have knock sensing. So if we are able to get this working, the possibilities are endless.

Same goes for the maf size, if we could figure how to convert tables (i'm sure piet will be able to find the correct adres) then we can run BMW 540 maf or bigger.

The 5 cilinder M44 is really great, and works really nice, almost as standalone but a bit more complicated.
 
I know I have seen it, great job! But now how to change them into some workable matter :)
But I'm sure Piet will be able to figure it out, he reads the bin like saturdays newspaper :D
 
I figured that out too :). Its just not in the version that is posted, because I am forgetful. I am pretty sure I copied the way it was done from the 5cyl XDFs, so it's probably his work already, lol. I'll try to update the first post.

1st Post updated with the version that shows the MAF tables correctly :).
 
Last edited:
Is there a solution to switch off rear lambda control?
Also what value do you guys use for load threashold lambda control?
Anybody know whats the load at peak hp?
One of the 204hp maps has better curve but still.
 
Rear lambda is not in the XDF? From my head it's called rear cat check 1, 2 and 3 they also have upper bank 2 and 3...

Abt load threshold lambda control you could make it a bit lower so that you start running rich a bit earlier, as soon as load threshold lambda control setpoint is reached it will run on VE table only. (by air mass ofcourse)

About load and how to read it depends of air mass and injector opening. Better to carefully read M44 wikia about the theoretical part.
Load 8 is some 240 hp in V70 T-5

If you want to run higher loads better to re-scale load scaling from 0 to 12.24
There is a re-scaling tool for this purpose. You could also find it on wiki.

Are you succesfully running and flashing M44 6 banger?
 
In the code, there will definitely be a difference. The addresses will be different. The maps are 'easy' to sort out, but most of the diagnostics are single-byte values, so there's no way to match patterns in the code and re-address the XDF for them.
 
Here are some addresses regarding the lamba probes / catalyst in the S90 bin:

1. "Rear Cat check (1)" (should better be called "Rear Lamba probe Heater check") : 0xEB0
2. "Front lambda probe heater check" : 0xEB1
3. "Rear Cat Check (2)" lower bank : 0x1025
4. "Rear Cat Check (2)" upper bank : 0x11025
5. "Rear Cat Check (3)" lower bank : 0x1137
6. "Rear Cat Check (3)" upper bank : 0x11137
 
I think for a 3.25 "ID MAF with the original S90 maf sensor in it, the MAF table should be multiplied by about 1.45.
I have made a small app that allows you to convert the MAF table by entering your own conversion factor.
Based on the air mass stationary, you can then tweak the conversion factor until the same air mass is measured as with the original S90 MAF..
In order to do that you need to be able to log, so I have added a log routine to the S90 bin.
And made an ADX for it.

I can't test it myself because I don't have an S90 (anymore).
So if anyone wants to test it, the files can be downloaded here:

-Bin with logging routine:
(deleted)

-the ADX that goes with it :
(deleted)

-S90 MAF conversion app for Windows:
(deleted)

A new improved set of files can be downloaded here:
https://www.dropbox.com/s/qw4khprm2odhe6m/S90 - Logging.zip?dl=0
 
Last edited:
Back
Top