Comparing "Desired Airmass" to Actual MAF Rate

[murfie]

And No that .85lb/min is not your desired air mass.

[CCS86]

It's the same idea that everyone has, correct your transfer function based on fuel trims if you don't have a know good transfer function. I just expanded on it saying it's not always your transfer function that's causing the error.

Above you did not say, "using fuel trims to change a transfer function introduces error."

You implied the flow rate reported will be changed directly by fuel trims, regardless of the source of the error. This isn't true from what I can see.

You seemed to be trying to poke holes in my [MAF Rate / Desired Airmass] concept by saying MAF rate isn't a reliable number. With an OEM housing, filter and transfer function it is.

Going off the values in your picture as an example, it is saying 1.03-1.04 lb/min is 17-18% more air than what was expected out of the transfer function. The ecu isn't going to use 1.04 LB/min to maintain lambda, it's going to use the corrected value ~0.85lb/min to get air mass to determine fuel mass from to maintain the proper ratio. Was the transfer wrong, was the fuel pressure wrong, was the fuel stoich wrong, was the sensor just reporting the wrong frequency, it could be anything number of factors. Hopefully you can take an educated guess and narrow things down.

It's not really "using a corrected airmass value". It is using the looked up airmass from the transfer function, right or wrong. Then it uses the O2s to modify pulsewidth, not the MAF signal.

In this case, the transfer function is untouched, the fuel system is stock, stoich AFR correct, etc. I'm working on this remotely, so I can't rule out mechanical.

[murfie]

If the MAF value was reliable, there would be no need for 02 feed back correction...

It's correcting the air mass value, you can see this in the air load PID.

[murfie]

The only thing wrong in the post you quoted today, was instead of "desired" I should have been more clear and said "predicted".

[murfie]

Idc about your MAF/desired MAF ratio concept. Understanding it, I know when I say "get your etc torque to match your engine brake torque", it will make this ratio 1. And unless you are not working towards a ratio of 1, we are saying the same thing.

[CCS86]

And No that .85lb/min is not your desired air mass.

I know. In that MAF lookup example, the "desired airmass" is 1.32 lb/min. It's a directly loggable PID.

If the MAF value was reliable, there would be no need for 02 feed back correction...

It's correcting the air mass value, you can see this in the air load PID.

I still don't understand why you associate fuel trims solely with "correcting airmass". Fuel trims are there to correct for all cumulative inaccuracies in the calculation and delivery of fuel.

There is always some combination of inaccuracy in how much air is entering the engine (vs measured/calculated) and how much fuel is being delivered. The sum total of inaccuracy is corrected via PW modification.

[murfie]

Just so you are aware, it can take up to 4-10 minutes before all startup/cold start open loop modifier tables are not influencing the fuel trims.