Anything described as dynamic means it is varying; it is in motion; it is changing over time. In an engine with an ECU that does calculations periodically, you can technically only know the past. Every sensor measurement has some time delay, and the algorithms those sensors inform, even more delay. Dynamic airflow exploits this fact to provide a good balance between accuracy of measurement and time-sensitive response. Past measurements of key sensors are compared with their current measurements and a determination is made about the airflow dynamics: it is steady state or it is in a transient.
A reference for this can be found here: https://patents.google.com/patent/US5423208A/en
The key sensors here are the MAP and throttle position sensors. The sensor outputs are monitored over time and the differences between subsequent measurements are evaluated against a set of steady state entry/exit criteria. In short, if the change from the previous measurement to the current measurement is over some threshold, steady state exits. If the inverse is true for some amount of time, steady state is entered.
dynamic.PNG
Loosely translated, when describing the source of engine airflow, steady state can be interpreted as MAF. Transient is speed density/VVE/VE. The logic here is that dynamic definition from earlier...things are moving. And during a transition, a MAF is several feet away from where you REALLY want your airflow/airmass measured: in the cylinder. And airflow isn't super quick relative to how fast the ECU is performing these measurements and calculations - so the MAF reading no longer directly corresponds to the real airflow entering the engine. In this case, the MAF measurement is ignored and the speed density modeled airflow becomes the source of air measurement for calculation of fuel requirements. This is depicted in the above image between steps 118 and 128. So long as the criteria are met and the MAF ignored, the MAP sensor with its close proximity to the cylinder (and other speed density parameters) is utilized to calculate fuel delivery. In this transient condition, the time sensitive nature of the speed density model produces the best estimation of airflow relative to what is happening in reality. Conversely, in steady state where the MAP and TPS activity have settled, time sensitivity is not as important and airflow calculation returns to the MAF. Given the MAF's insensitivity of measurement to barometric pressure, temperature and exhaust pressure, it is considered the 'source of truth' by which the speed density/VE is corrected...so long as it remains in steady state.
A reference for this particular VE correction algo is here: https://patents.google.com/patent/US5465617A/en
VEcorr.PNG
It should be clear here that dynamic airflow can and does produce airflow values from both the MAF and speed density model. Again, see steps 124-128 in the first image. Which one it matches will depend on those steady state criteria. It should also be obvious that, since you don't know the source of the airflow based on dynamic airflow alone (especially since the steady state enable criteria are not defined by HPT in any gen 4 ECU calibration I have seen...see gen3 for examples), it should not be used as reference for anything related to speed density/VE/VVE or MAF calibration. I have purposely excluded the prediction contribution of dynamic airflow from this overview - however that presents another reason that dynamic airflow, despite being the source of air mass for fuel calculations, does not represent either the speed density model or the MAF and thus should not be used to correct them.
Reply With Quote
Originally Posted by smokeshow
