closed loop operation

[ike84]

When the vehicle is operating in closed loop operations,

Does the MAF provide the main input which is then fine tuned by the NBO2 sensors?

OR

Does the NBO2 sensor provide the main input when is then fine tuned by the MAF?

Logic would dictate that it is the former but I needed to ask if anyone knows for sure.

[blue62]

Originally Posted by ike84

When the vehicle is operating in closed loop operations,

Does the MAF provide the main input which is then fine tuned by the NBO2 sensors?

OR

Does the NBO2 sensor provide the main input when is then fine tuned by the MAF?

Logic would dictate that it is the former but I needed to ask if anyone knows for sure.

Short answer:
MAF Which includes Intake Air Temp, TPS, coolant temp, provide the main input.

O2 sensors provide feedback =(Closed loop) info. in the form of voltage to the DME/ECU as confirmation as to what the DME/ECU has already done.
They act as a witness to an event. Yes your going lean or yes your going rich.

The actions of the DME/ECU drives the O2 signal. That is why you see the voltage signal of an O2 sensor as a sine wave. The DME/ECU drives the O2 signal high (rich) then low (lean). Then high, then low, then high, then low. constantly.

[ike84]

I just wanted to make sure that it is the nbo2 sensors that, as you said, serve as a witness to the MAF data and not the other way around.

Correct me if I'm wrong, but our cars will still operate with the maf disconnected, which is why I questioned that.

You don't happen to know the trigger/threshold voltages on the nbo2 sensor readout that actually causes the fueling changes, do you? I've studied the different voltage vs AFR plots and the curve's temperature dependence, but I haven't been able to find any hard number about the ecu's thresholds for reversing the fuel trend.

The below pic came from a Bosch pdf, but they have inconveniently omitted the axis scales. The other chart is an actual plot, but no info on what sensor was used or at what temp. My best to guess is .7 and .25 are the thresholds based on overlaying trust two charts.

Can you guess why a tinkerer such as my self would be interested in this info? I'll give you a hint, there's more power at 14.2 than there is at 14.7




Sent from my SM-G970U1 using Tapatalk

[blue62]

Originally Posted by ike84

I just wanted to make sure that it is the nbo2 sensors that, as you said, serve as a witness to the MAF data and not the other way around.

Correct me if I'm wrong, but our cars will still operate with the maf disconnected, which is why I questioned that.

You don't happen to know the trigger/threshold voltages on the nbo2 sensor readout that actually causes the fueling changes, do you? I've studied the different voltage vs AFR plots and the curve's temperature dependence, but I haven't been able to find any hard number about the ecu's thresholds for reversing the fuel trend.

The below pic came from a Bosch pdf, but they have inconveniently omitted the axis scales. The other chart is an actual plot, but no info on what sensor was used or at what temp. My best to guess is .7 and .25 are the thresholds based on overlaying trust two charts.

Can you guess why a tinkerer such as my self would be interested in this info? I'll give you a hint, there's more power at 14.2 than there is at 14.7




Sent from my SM-G970U1 using Tapatalk

First:
O2 sensors "do not" serve as witness to MAF data.
They serve witness to what the DME is doing with fuel.... in an indirect manner.
O2 sensors compare oxygen in the exhaust to oxygen in the atmosphere.
From that comparison a small voltage is produced. That is the only thing they are capable of.

Yes our cars will run with the MAF unplugged.
The DME will default to open loop and use default fueling maps.

Threshold voltage:
Stoichiometric= 14.7-1 fuel/air ratio "by weight" which is represented by .450mv on a NBO2 sensor.
Above .450mv is rich
Below .450mv is lean
In simple terms what happens is:
From a precaluculated set point "14.7-1 fuel/air" on a fuel map.
The DME will add fuel until the O2 sensor voltage crosses the .450mv. "Threshold"
It will then start taking away fuel until the O2 sensor voltage falls and again crosses the .450mv "threshold".
Then it will start adding fuel again to drive the O2 sensor voltage back up across the
the .450 threshold.
So on average the fueling strategy is Stoichiometric. 14.7-1 Fuel/air
Depending on the manufacture, model, year, etc. etc.
This happens around 3-10 times per second.

The above is in very simple terms.
What is happening is why I say the O2 sensor just bares witness to what the DME is doing and that the DME drives the O2 sensor signal.

The O2 sensor voltage range of the sensor used in our cars ranges from around .100mv - .950mv

I look at all sensors as witnesses to an event and the DME/ECU as the Judge.
The witnesses report to the judge and the judge makes engine management decisions based on the witnesses testimony.
Like all witnesses they are either telling the truth. (So the sensor is not the issue in relation to a fault code). Or they are lying...(faulty sensor).
Thinking this way is just part of my diagnostic methodology.

Yes I am well aware of fuel air ratios and their affects on power.

If your wanting to tweak fueling and and watch AFR's you want to install wideband lamda sensors. They actually sense fuel air ratios not just oxygen.
NB O2 sensors are only accurate at Stoichiometric 14.7-1 AFR so very poor for what your looking to do.

The number one reason the DME/ECU has the fueling strategy it has... is so the Catalytic converter performs properly and has long life.
The second reason is fuel economy.
What you and I call performance falls third.
The first two things are mandated by federal law
The third.... Performance... is just a bonus

Your graph on the left (the brown one) is voltage "mv" on the left and fuel/air ratio on the bottom
Look across the bottom tell you find 14.7 go up and the O2 sensor signal will be intersecting .450mv on the scale at the left.

[ike84]

So rather than there being an operating range with high and low ends, the dme is just looking at greater or less than .45v? That makes my project much simpler! Thanks blue.

Sent from my SM-G970U1 using Tapatalk

[blue62]

Originally Posted by ike84

So rather than there being an operating range with high and low ends, the dme is just looking at greater or less than .45v? That makes my project much simpler! Thanks blue.

Sent from my SM-G970U1 using Tapatalk

Yes in simple terms that is sort of correct.
There is an operating range of both the O2 sensor and the DME
What I described is (as I say in very simple terms) how the STFT strategy works.
The primary reason for the rich-lean rich-lean strategy is Catalytic converter function.

It would be way better if you did some research on how the various sensors work and their relationship to each other.
Explanations can be rather complex.

If you want a better understanding of how the basic fueling strategy on a modern car works and it's relationship to the overall OBDII system. You need to study up on the function of:
MAF sensors, O2 sensors, and Catalytic converters, and most importantly their relationship to each other.
For MAF and O2 sensors research via "Bosch" they developed both for automotive use.
Also do a Wikipedia search for Catalytic converter function. Pay close attention to the section on three way Cats. That is what is used on all U.S. sold cars.
OBDII and fueling strategies are built or developed around Catalytic converter function.
Three way Catalytic converter function in particular.

Once you get a handle on how the sensors and cats function then go to youtube and look up "Scannerdanner" (Paul Danner) study everything he has on fuel trims. check his website out. he also has a book or two in print.

Do all I have mentioned and you will start to have a "general" understanding of how fueling and the OBDII system on a modern day car works.

Another interesting subject to look at is:
Rich burn versus lean burn engines....

[ike84]

I'm not an expert but I've exhausted my ability to learn about the sensor function operations so far, my questions are really toward how ME7.2 manages the information.

My plan is to modify the nbo2 sensor readings back to the dme in order to run rich.

The reason i asked whether the maf checked the nbo2 or vice versa is because the explicit logic that me 7.2 uses is not described anywhere that I can find and it will impact the design of my system.

If the NBO2 is the "trump all others" measure of fuel mixture settings in closed loop operations, then I can create a voltage divider to accomplish this task.

If the NBO2 readings are of secondary importance, then I need to focus moreso on modulation of the MAF output with an opamp and just feed back a constant .45V to the nbo2 sensor pickup at the dme.

The second approach would be less complicated but I don't think that it would work if the computers logic is centered around a [greater than or less to 0.45V execute x func] since the dme REQUIRES fluctuations, as opposed to a system that is shooting for a specific operational value range that it is constantly adjusting the variable outputs (i.e. fueling) to stay within.

[blue62]

Originally Posted by ike84

I'm not an expert but I've exhausted my ability to learn about the sensor function operations so far, my questions are really toward how ME7.2 manages the information.

My plan is to modify the nbo2 sensor readings back to the dme in order to run rich.

The reason i asked whether the maf checked the nbo2 or vice versa is because the explicit logic that me 7.2 uses is not described anywhere that I can find and it will impact the design of my system.

If the NBO2 is the "trump all others" measure of fuel mixture settings in closed loop operations, then I can create a voltage divider to accomplish this task.

If the NBO2 readings are of secondary importance, then I need to focus moreso on modulation of the MAF output with an opamp and just feed back a constant .45V to the nbo2 sensor pickup at the dme.

The second approach would be less complicated but I don't think that it would work if the computers logic is centered around a [greater than or less to 0.45V execute x func] since the dme REQUIRES fluctuations, as opposed to a system that is shooting for a specific operational value range that it is constantly adjusting the variable outputs (i.e. fueling) to stay within.

A few very limited things I can tell you is.
Motronic or the DME/ECU was as you may already know developed by Bosch.
I have only found limited info on the inner workings.
Searching everything I could find via Bosch only gave me very limited info.
The Motronic DME/ECU in it's various forms has been used in a wide range of automobiles.
Porsche, Rover, BMW, and even GM early on.

What you might try is cruising the BMW fourms those guys seem to be more informed and have dug deeper into the workings of the Motronic then any other place I have found.

There are people within the BMW community that have attempted or done the sort of things your talking about.
Tweaking fuel maps and lookup tables as well as changing or attempting to change the programing of the DME/ECU.
Perhaps you could find some useful info via the BMW guys.

[ike84]

I found this on a bike forum page. Looks like the dme is looking for a range and not a fixed point, which may make this a lot simpler!

If this is the case, I be should be able to feed .45V back to the dme and effectively disable the feedback of closed loop operations! Then I can control the air fuel mixture in both open and closed loop operations with a signal amplifier on the MAF sensor.

Here's my real question though - why has no one else already done this?

Sent from my SM-G970U1 using Tapatalk

[blue62]

Originally Posted by ike84

I found this on a bike forum page. Looks like the dme is looking for a range and not a fixed point, which may make this a lot simpler!

If this is the case, I be should be able to feed .45V back to the dme and effectively disable the feedback of closed loop operations! Then I can control the air fuel mixture in both open and closed loop operations with a signal amplifier on the MAF sensor.

Here's my real question though - why has no one else already done this?

Sent from my SM-G970U1 using Tapatalk

Yes they are saying the same thing I described.
Bottom line is the DME/ECU is attempting to create an average of 14.7-1 AFR
or Stoichiometric.