Originally Posted by
BenFenner I'll offer up an explanation for why the TP and torque trace shapes don't match as well as they should.
The TP trace is the estimated (theoretical) injector pulse width needed to provide enough fuel to produce a stoichiometric mixture. The incoming amount of air is read by the MAF and this is an estimation of power. The RPM is used in the TP calculation to turn this estimation of power into an estimation of torque. The torque is then treated to a scaling factor (the K value) based on injector size which results in the theoretical pulse width (TP) number.
(MAF VQ / RPM) * K value = TP
(estimated power / RPM ) * injector flow-rate scale factor = TP
(estimated torque) * injector flow-rate scale factor = TP
This is the theoretical amount of fuel needed for a 14.7:1 AFR at a given RPM and injector size. As you should know by now, injector pulse width on a properly tuned engine for a given AFR will follow the torque curve. More air and fuel for a single combustion event means more torque. Less air and fuel for a single combustion event means less torque. Assuming a proper ignition timing setup.
For this reason, TP can be thought of as the ideal torque curve based on the available air as read by the MAF sensor, and scaled by RPM and the K value. The actual torque curve measured on a dyno should be very, very similar once you have the fueling and ignition timing down correctly. It can be seen that on the first of Vadim's examples he's missing a good bit of torque between 4,500 RPM and 6,500 RPM. He obviously has the air to work with, as evidenced by the TP readings, but for whatever reason he is not making use of it efficiently.
The same goes for the second graph where you can see early on at 3,400 - 4,500 RPM the engine is not making efficient use of the available air. And again there is 5,200 - 6,200 RPM where there could be higher torque numbers if it weren't for some issue in the tune. The air is there, but the measured dyno torque does not follow for whatever reason.
And once again, for the final time, TP does not mimic MAP readings.
BenFenner you’re trying to make a connection between TP and Torque, when there is none.
Seem like you got torque out of the blue. There is no torque input in Nissan’s TP calculation. You are artificially manufacturing one.
The TP is
not an "estimated/theoretical injector pulse width to provide enough fuel to produce a 14.7 afr". TP is more basic than that. It’s only a injector pulse width in milisecond. It might or might not hit 14.7 afr, it all up to you. You’re the one adding or subtracting the TP/ injector pulse width in order to hit 14.7 or whatever afr you want.
TP = (MAF VQ x K Value/ RPM) + injector latency and enrichment
The final product of the TP formula = injector pulse width in miliseconds.
Like I said before, TP does not follow the torque curve period. Anybody with map trace capability can see this when they do a full throttle pull.
I thank you and Vadim for providing the raw data logs and excel graphs to illustrate my point. TP does not follow the virtual dyno torque curve. If you think TP will follow the torque curve of a real dyno more closely , you’re going to be disappointed. The difference between the two will be even greater.
As we all can see, when you try to put TP on the same graph as the torque, the TP curve will be flat as a witch tit.