I follow you Zig. There's nothing tricky about the way the model works. It works as you would expect it to. The ailerons change the effective AoA of that portion of the wing and the resultant forces generate the effect. Looking at a typical roll rate chart for these planes, roll rate will increase with speed up to an apex and then fall off as stick force becomes a limiting factor. Once the data is plugged into the model, the plane should be closely match the part of the chart to the apex. If it doesn't then that tells me that something is off and I have to massage the variables to produce the desired output. Once you get past the optimal speed, roll rate is decreasing because maximum stick deflection can no longer be achieved. At that point, I'm looking at the backside of the roll rate curve and using that to input to the model that maximum lateral stick movement for that speed. In this case, the inputs for limiting stick deflection are solely based on the desired output. It is a lot of inputs to make and any mistakes made there are not readily apparent. It also means that if I later find something amiss on the front side of the curve and correct it, it throws everything off on the back side. One of the changes that I alluded to is that we're changing the input to the other side of the equation. Instead of stating what the maximum stick input is at a particular speed, the maximum roll rate will be used instead and the model will fill in the necessary stick input. That will give us a higher degree of accuracy and make errors less likely. We're also making changes to get rid of some limitations that introduce small errors that have to be overcome elsewhere. An example of this is that the current model assumes aileron travel is equal in both directions. That's not the case on many planes and modeling it that way could lead to some minor undesired characteristics.
As HT said, moments of inertia are nice to have but not something we really sweat. We had a good starting point and having an exact number is not needed because even a huge change can be barely perceptible. You really just need to get in the ballpark with it.
I don't know why some people feel there should be a tremendous amount of roll inertia in these planes. For me to impart a large amount on a typical single engine fighter, I would have to increase the MoI's an order or more of magnitude. It was something that we talked about prior to HT taking his P-51 flight because I thought that it should be present to a noticeable degree(I'm not talking about a big bellybutton lag, but rather the absense of crispness). Anyway, his flight laid those doubts about the model to rest. I guess the reason some people subscribe to that idea is because it seems to make sense intuitively. It's much easier to visualize mass than force.
