Sim Throttle Response Curves Explained
iRacing just release their new build to rave reviews on the forums. One of the changes was the throttle response curves for all the cars. It's been a very subtle, but fantastic change, but understanding exactly what happened is tricky.
Here's Grant Reeves explanation...
==========================
The old throttle response curve was a simple y = x^2, which was done like a million billion years ago, way before I ever arrived at Papyrus in 98 and no one has ever thought to go back and look at it. We would know for all those cars we've modelled for the last 15-20 years what the full throttle dyno curve looks like, and what the off throttle engine braking looks like, but how the engine responds in between those two extremes I guess was never realized to be quite an important driving characteristic.
I believe the original logic was that y=x^2 approximates the shape of a 1-cosine curve from 0 to 90 degrees which was supposed to vaguely model the opening of the butterfly valve in the throttle body (but in my mind I'd expect that the butterfly valve opening would be better modelled with a sine curve than 1-cosine, and that would have been a far better match to the actual response)
The new throttle response curves are based on several sets of real vehicle data. One of the cars we model even has dyno information for full rpm pulls at every 10% of throttle position. By comparing how the engine power at any given rpm varies with throttle application I came up with an approximating function that sort of looks like the average of all the data we have.
At very small throttle openings, the airflow is pretty much choked to a fixed amount, so regardless of engine rpm it's only getting a fixed amount of air to burn for power output. Thus the initial (first 10-15% or so) throttle response regardless of rpm is kind of proportional to power output.
But at full throttle, the power output is limited by how much air the engine can pump. At low rpms the engine can't flow much air so it very quickly saturates it's power output, which is why at low rpm you feel like you're getting 80-90% of engine power by 40-50% of throttle, and the rest of the throttle travel feels very unresponsive. But at high rpm the motor can pump a whole lot of air, so the power output keeps increasing the further you get into the throttle.
Eventually you start running into limitations of how much air the throttle body can flow when wide open, and into limitations of the engine's volumetric efficiency at pumping air at very high rpm, which causes the throttle response to still flatten off as you get towards 80-90% even at max rpm.
It's quite likely that every engine in existance has subtley difference throttle response based on various build characteristics, but we're just giving them all the same response curve for now based on what their rev limit is.
First time I tried the new curves - like most of you guys have noticed - it
immediately felt incredibly more natural - even the Spec Racer (which I like to call a Death Trap) was suddenly a whole lot more controllable and drivable. So, yeah, it does seem that it's pretty important to model what happens in between 0% and 100%.
========================
When iRacing decides to update something they really put a lot of thought into it. I can't wait to see their new transmission model they're set to release in the next 30 days.
Great stuff!