Desired load

[mnbf69]

I've searched this forum top to bottom to try and find a reason as to why my desired load never goes higher then ~1.11. my air load is 1.3+
car is boosted on a control pack gen 2 manual mustang ecu. I've tried to modify my inverse torque and driver demand with no luck.
if anyone could point me in the right direction it would be much appreciated.
uio.hpl

[Pistol_91]

I also wonder about this

[engineermike]

What does your max map over Bp table looks like?

[mnbf69]

no.jpg

[engineermike]

Ok that’s not it. What’s your etc torque request look like at that time?

[mnbf69]

I uploaded my log in my first post

[veeefour]

Desired load is the max value taken from torque model. Speed density max load has nothing to do with this.

OP post your file or shots from torque model.

[mnbf69]

Desired load is the max value taken from torque model. Speed density max load has nothing to do with this.

OP post your file or shots from torque model.

This is just the base start point so ignore the rest of the tune this is just for torque model reference

[veeefour]

Well confirmed, your model covers values up to 1.5 only but not only that but that answers your main concern.

Anticipation and initialization is on while SD model leaves a lot to be desired. MAF looks stock. Amigo this not even close to a starting point.

[veeefour]

The easiest way to extrapolate the torque model is to take what you have and multiply times 2.

You need 0.1 and 0.3 maybe 0.7 and then jump to 1.0 as boost will jump over vacuum area quite fast and you dont really need that range anymore.
No take values from load 0.8 multiply by two and yo get a good starting value for load 1.6. Remove some rows under load 1.0 and fill boost range like that.

Torque model is linear, this will place you in a good starting point of course not perfect but you got the idea.

Second part is a request from Driver Demand.

[mnbf69]

So bacicly your saying because anticipation and initialization is on is why it is not going over 1.1 desired load?
I know the rest of the tune is not in a good place in this specific tune but like I said i just posted it to show the tq model

[veeefour]

Its not going anywhere because your torque model is not accurate, neither torque request.

The other problem is messing with your air model so its affecting the torque calculations.

[mnbf69]

I appreciate you taking time to explain as I'm pretty new to tuneing, I've been trying to use many differnt stock roush OS tune models and all have had the same result with the desired load, the only thing I did not know about was the anticipation and initialization being a factor. I realise once it is working correctly that the tq modle will need some fine tuning.

[mnbf69]

The easiest way to extrapolate the torque model is to take what you have and multiply times 2.

You need 0.1 and 0.3 maybe 0.7 and then jump to 1.0 as boost will jump over vacuum area quite fast and you dont really need that range anymore.
No take values from load 0.8 multiply by two and yo get a good starting value for load 1.6. Remove some rows under load 1.0 and fill boost range like that.

Torque model is linear, this will place you in a good starting point of course not perfect but you got the idea.

Second part is a request from Driver Demand.

Untitled.jpg does this seem like what you are referring too?

[veeefour]

More less yes, but try to keep equal spacing between rows and columns. Keep it smooth. Use stock values for low load.Use VCM built in cal to take care of Inverse.

Below the example how a good torque model for high boost should look like:

image.png

[mnbf69]

More less yes, but try to keep equal spacing between rows and columns. Keep it smooth. Use stock values for low load.Use VCM built in cal to take care of Inverse.

Below the example how a good torque model for high boost should look like:

image.png

So I'm curious as to why a stock roush tunes tq model would not work for me? does it have to do with the OS differences? i know ive run the tq model through the VCM calculator to get the inverse and the numbers don't match the stock roush inverse numbers.

[engineermike]

The Roush stock torque model is pretty good and will often work.

Your desired load tops out a little over 1.0 because your driver demand is only ~400 ftlb. Look under torque management\driver demand\desired torque\normal and you'll see that even at 90% pedal (max) you're only requesting 300-400 ftlb. You'll need to increase these values to get torque request much over 1.0. The way the logic works is torque demand -> torque to load -> load @ rpm is converted into airflow -> airflow requirement and TB dp is used to determine TB angle needed to achieve airflow via the TB model -> airflow PID feedback loop using MAF is used to correct error

FYI the 44351 table can also limit driver demand, as can predicted injector pulsewidth and other things. But in your case, the most obvious issue is in the torque demand tables.