Not really sure, I would need to compare SD based os for 6r80 how it differs from what I used as base. And probably it would be easier just to port my changes onto that OS.
What is use case where you would want to switch from MAF to SD ?
Im generally asking cause my twin turbo car with 6870 twins is doing just fine still on MAF and I just don't seem to feel where SD would be better/easier.
I have in mind adding nitrous support into LisevOS since I figured all I would need to do is simply add some airflow into MAF sensor function and it would increase needed fuel and increase airload which would handle spark related stuff easily
Not really sure, I would need to compare SD based os for 6r80 how it differs from what I used as base. And probably it would be easier just to port my changes onto that OS.
What is use case where you would want to switch from MAF to SD ?
Im generally asking cause my twin turbo car with 6870 twins is doing just fine still on MAF and I just don't seem to feel where SD would be better/easier.
I have in mind adding nitrous support into LisevOS since I figured all I would need to do is simply add some airflow into MAF sensor function and it would increase needed fuel and increase airload which would handle spark related stuff easily
As far as use case. For max effort NA builds, the MAF restriction could be meaningful. I do want try an OTR which is apparently tricky for MAF to work well in that quick ram up into the TB. And I suspect transient fueling and feel may be improved with a MAP only calibration.
As far as use case. For max effort NA builds, the MAF restriction could be meaningful. I do want try an OTR which is apparently tricky for MAF to work well in that quick ram up into the TB. And I suspect transient fueling and feel may be improved with a MAP only calibration.
I've helped my friend build max effort 10r80 N/A ( 700 rwhp ) with 5" diameter OTR I've welded up from aliexpress pipes and bellmouth and only problem I had was getting idle to work properly in windy conditions ( wind was blowing through MAF sensor and bouncing back from closed throttle ), other than that it worked without any problems or differences from Steeda CAI, at least in drag racing/dyno environments.
And that idle condition can probably be tuned out since I found there is MAF smoothing time table based on engine rpm ( with value of 5ms for 1000 rpm and going up to 20ms for 6000 rpm ).
But in case if you wanted to switch to MAP settings, from hardware point of view, MAP is just analog sensor that takes 1 pin from ECU. IMRC1 and IMRC2 plugs have 5V, GND, and ecu PIN and as far as I know Roush OS just uses those pins. On gen2 those are Molex MX150 plugs and making harness would be simply making extension that you plug into IMRC and MAP sensor on other end.
On my car, I use IMRC1 / IMRC2 just to log Fuel pressure and MAP.
F-150 gen2 calibrations have separate MAP pin that is unused on Mustangs
Not really sure, I would need to compare SD based os for 6r80 how it differs from what I used as base. And probably it would be easier just to port my changes onto that OS.
What is use case where you would want to switch from MAF to SD ?
Im generally asking cause my twin turbo car with 6870 twins is doing just fine still on MAF and I just don't seem to feel where SD would be better/easier.
The use-case for NA isn't as strong as SC or turbo but still exists. The Ford logic is determining the properties of the air in the plenum to then figure out what the properties of the air in the cylinder are on a per-cycle basis. With MAF in steady state, this is easy. The mass of air that passes through the MAF sensor is the same as what makes it into the cylinder. load=MAF/rpm/4/.0017. However, in dynamic situations, it does some very complex math where it figures a certain amount of air is entering the plenum (from the MAF sensor) and a certain amount of air mass is leaving the plenum and entering the current cylinder on intake (using SD model) so it then knows how much mass of air remains in the plenum after the current cylinder, therefore it attempts to predict how much air is entering the next cylinder so it can then know how much fuel is needed. You can see how this is less-than-ideal with lots of room for error. Will it work? Of course, but it can be better. As a side note, Ford did finally switch the Mustang GT over to SD in the S650, so I'm not sure if that leaves any vehicles using MAF in their lineup. SD, as you know, directly measure pressure and temp in the plenum and, therefore, knows from the SD model how much air will enter the next cylinder without needing all the prediction math.
With PD blowers, there is another problem. Ford's throttle control needs to know the TIP (throttle inlet pressure, assumed to be baro in the Mustang GT) and the TOP (throttle outlet pressure, assumed to be inferred MAP in the Mustang GT) to know what angle to send the blade to in order to achieve the desired air flow rate. With NA and PD blower with bypass open, the TOP = MAP, so all is well. However, when the PD bypass closes, MAP no longer equals TOP. So with MAF it doesn't "know" what TOP is and throttle control is compromised. In most situations it doesn't matter much because a) we don't generally throttle the air while in boost and b) most aftermarket tunes open up the torque error tolerance to hide the problem. However, for instance, Roush and GT500 both have MAP *and* TOP sensors, with the TOP sensor there for throttle control and MAP there for the SD airflow and fueling logic. This is why I am personally intersted.
Beyond that, you may have noticed above that the Coyote logic assumes TIP to be baro. This presents a problem with centrifugal superchargers and turbos. I believe ecoboost adds a TIP sensor, again, to be used for throttle control when TIP is not baro. It would be better if turbo or centrif Mustangs could use some of the ecoboost code and add the TIP sensor too. Again, not that this causes a problem because they just open up the allowable torque errors to hide the problem.
@geddys according to Banish, Ford uses global logic meaning all cars have basically the same code, just with features and functionality turned on or off and different calibration data. For instance, a Ford Maverick and everything else has the code for burbles, but it's just turned off. I'd bet lunch that the SD code exists in the mustang coyotes but there is a hard-coded "switch" or definition exists where you can turn MAF off and SD on.
@geddys according to Banish, Ford uses global logic meaning all cars have basically the same code, just with features and functionality turned on or off and different calibration data. For instance, a Ford Maverick and everything else has the code for burbles, but it's just turned off. I'd bet lunch that the SD code exists in the mustang coyotes but there is a hard-coded "switch" or definition exists where you can turn MAF off and SD on.
They do use one master code for all of the engines, however once they tailor firmware for specific vehicle engine/trans/sensors combinations they remove unused big features from code base ( probably there is a lot of #ifdef in codebase ).
For example I found multiple functions that read voltage for every pin in ECU that has ADC support but some of them are never called anywhere since logic for specific feature was stripped during build time.
In GT500 calibration for example, DI logic is completely stripped out, even though there are still some artifacts with some DI parameters left in firmware.
Thats the reason you need to compare directly two firmwares to determine is certain feature is still there, just disabled or maybe some of logic is there and just sensor reading is missing.
About SD vs MAF, thanks for insights. I guess none of scenarios you mention really apply to my car and probably that reason I never needed it. My turbos are big enough with quite big cam that going part-throttle prevents them from going into meaningful boost so I never had that issue with controlling car just cruising around. And boost is controlled externally so ECU just tries to match up with needed fuel. I did notice that tuning SD maps properly did improve driveability
