AAT Airflow Meter: A Thought Experiment

Hi,
What would actually happen if you were to jam the flap in an Audi AAT mass airflow sensor in the "fully open" position?
The engine control unit (ECU) would then be constantly receiving a signal indicating the maximum possible amount of air, even though that wasn't actually the case.
What happens next?
It should be similar to the diode trick used with the 1Z/AFN.
(AGR is not present in this thought experiment.)

Regards,
Jens.

Jens 16syncro wrote:

Hi, What would actually happen if you were to jam the flap in an Audi AAT air mass meter to the "fully open" position? The engine control unit (ECU) would then be constantly receiving a signal indicating the maximum possible amount of air, even though that wasn't actually the case. What happens next? It should be similar to the diode trick used with the 1Z/AFN. (AGR is not present in this thought experiment.) Regards, Jens

That's correct; it would simply mean there would be no more restrictions on smoking.

However, using an EGR system in this way would cause too much exhaust gas to be recirculated, as the EGR valve would be fully open at all times, even under full load.

Let's explore this thought experiment further.
The N75 valve is bypassed in the AAT system and replaced with a mechanically adjustable valve.
This is configured so that the wastegate opens sufficiently when the maximum boost pressure is reached. This essentially resets the turbocharger to a "fully mechanical boost limiting" state. The turbocharger then only has the internal wastegate control loop.
The engine control unit (ECU) has a boost pressure map, and parts of this map would not be achieved because the external control system is no longer active.
What happens then? Does the engine control unit (ECU) go into a limp mode? Or is the current boost pressure indicated by the boost pressure sensor simply converted into signals for the pump's flow control, regardless of whether it corresponds to the target values?

Regards,
Jens.

In this context, the position of the mass airflow sensor primarily affects the air mass, but depending on the engine, there may be protective functions in place to prevent excessive boost pressure, either by entering a limp mode or by reducing the fuel injection quantity.

With lower boost pressure (i.e., below the specified range), a limp-home mode can also occur, but it depends on the engine and control unit. In my case, however, there is no limp-home mode if the boost pressure is too low due to a disruption in the electrical connection to the N75 valve.

Regarding the turbocharger, you would experience a significant turbo lag because the stock unit only achieves approximately 0.6 bar of boost. If you were to use a different unit capable of producing 1 bar, you would also need to incorporate safety margins to protect the turbocharger.

Malte1408 wrote:

The location of the aggregate plant primarily affects the air mass

.

Hmm, I don't believe that.
The air mass cannot be used as the sole control parameter. The 1Z engine, for example, can function even without an air mass sensor, as it can simulate a maximum air mass using a diode.
Influencing factors should rather be:
Pedal position sensor.
Engine speed.
Boost pressure.

The Audi 1T, for example, does not have a mass airflow sensor as standard equipment.

Regards,
Jens.

christians wrote:

Regarding the boost pressure issue, you would experience a significant turbo lag because the stock turbo only achieves approximately 0.6 bar, and even with a different turbo that could reach 1 bar, you would need to incorporate safety margins to protect the turbocharger.

Okay, I would leave the can as it is. The hose from the intake manifold (boost pressure) to the wastegate should also remain, but a throttle should be installed in the hose so that the actual maximum boost pressure (around 1 bar) is achieved.
Then the turbo is regulated in a purely mechanical way, just like in an old swirl chamber turbocharger.
The turbocharger's performance then depends solely on the geometry of the turbocharger itself.
You don't necessarily need to build in "safety margins"; instead, you should simply limit the maximum boost pressure to a reasonable value, where the wastegate will open accordingly.

Hello Jens,
Unfortunately, I don't have the RLF (Release Log File) for the AAT, but if I remember correctly, the old system already had that feature.
'Charging pressure regulation deviation' -> emergency mode.


Best regards,
Rudi

Quote:

The 1Z engine can also function without a mass airflow sensor, if you simulate a maximum air mass using a diode.

Then the air mass will still be used for the calculation, but it will no longer be a limiting factor.

If the application doesn't have an LMM (Linear Motor Module), then the LD (Linear Drive) is actually used instead, but it's a different design. That's why a 1Z with the LMM disconnected won't function the same way as a 1T.

The process of determining the injection quantity works as follows:

1) There are 3 calculated injection quantities:

Driver request, derived from pedal position and engine speed, possibly including speed.

Torque limitation, calculated from speed.

Smoke suppression, consisting ONLY of air mass, which is the air mass indicated by the TDI (Turbulence-Dominated Instability) measurement.
(mg/dose) ALREADY INCLUDES the speed, which is calculated from the mass airflow sensor voltage and the...
Calculated the speed.

The smallest of the three injection masses is then injected, UNLESS the engine is idling or operating at its maximum speed. The cruise control is activated based on the driver's preference.

There are also a number of safety and comfort features, such as vibration control and shock absorbers, that process additional data to optimize the system's performance. I'm not entirely sure exactly how these routines work. (I suspect they function in a similar way to the idle speed control systems.)

Okay, if the injection solution now contains 10mg per vial, they need to be adjusted on the ventilator.

Ultimately, the amount of fuel injected by the VP depends on the position of the throttle, the engine speed, and the fuel temperature.
The throttle valve has a position sensor, and the fuel temperature is measured.

Based on the required fuel injection volume and fuel temperature, the control voltage for the control valve position sensor is determined from the pump characteristic curve. This voltage corresponds to the desired position of the control valve. This position is then set by the actuator. (Position control)

"Those who understand the principle will realize that the amount of fuel injected is not regulated, but only controlled. Furthermore, the amount of fuel injected is no longer being measured, and for good reason..."
The location of the signal box is the issue.

One could also say that the position of the control lever is the actual controlled variable, while the injection mass is the manipulated variable, which is linked to the controlled variable via the pump characteristic curve.

Malte1408 wrote:

1) There are 3 calculated injection quantities: Driver request, derived from pedal position and engine speed, possibly including speed. Torque limitation, calculated from speed. Smoke suppression, consisting ONLY of air mass, which is the air mass indicated by the TDI (Turbulence-Dominated Instability) measurement. (mg/dose) ALREADY INCLUDES the speed, which is calculated from the mass airflow sensor voltage and the... Calculated the speed. The smallest of the three injection masses is then injected, UNLESS the engine is idling or operating at its maximum speed. The cruise control is activated based on the driver's preference.

Thank you for the explanations.

If only these three parameters determine the fuel injection amount, then the boost pressure is completely missing. The boost pressure would not be a control variable for the fuel injection amount.
Therefore, the boost pressure map is a consequence of other factors; in other words, the boost pressure to be set is determined by the engine speed and (calculated) fuel injection quantity (based on three maps).
If the control unit detects significant deviations from the expected boost pressure curve, what happens?
=> Protection function activated due to any deviation? (Error 575: Intake manifold pressure difference)
=> Protection function only activated when a maximum value is exceeded?

Jens.

Quote:

If there are only these three parameters that determine the injection amount, then the boost pressure is completely missing. The boost pressure would not be a control variable for the injection amount. Therefore, the boost pressure map is a consequence of other factors; in other words, the boost pressure to be set is determined by the engine speed and (calculated) fuel injection amount (based on three separate maps).

Hmm, yes, I would phrase it differently.

First, the "regulated variable" is the variable that is being controlled. This means that boost pressure is already a regulated variable, but it's the regulated variable *of* the boost pressure control system.

However, it is not a control parameter (setpoint or influence on the setpoint) for the fuel injection quantity control. The boost pressure affects the fuel injection quantity control through the resulting air mass. (No boost pressure -> No air -> No fuel).

The target boost pressure is a control parameter for boost pressure, derived from the boost pressure map based on engine speed, actual fuel injection mass, and intake air temperature (for protection).

It becomes a bit of a chicken-and-egg problem: the entire mapping needs to be tuned in such a way that there's still a reserve capacity to increase the fuel injection amount. This, in turn, generates more boost pressure, which allows for an even further increase in the fuel injection amount.

The emissions regulations, for example, stipulate that the opacity must not exceed a value of 2. Beyond that, the smoke curve determines the engine's response characteristics, including factors like turbo lag, etc.

Quote:

If the control unit detects significant deviations from the expected boost pressure map, what happens? => Protection function activated for any deviation? (Error 575: Intake manifold pressure control difference) => Protection function only activated when a maximum value is exceeded?

I don't know, it can vary depending on the software. There's probably an error log entry, and a limp-home mode can also be triggered by insufficient boost pressure, but that was an AFN engine with a VTG turbocharger.

The easiest way to try it out.

Malte1408 wrote:

"] Quote: If the control unit detects significant deviations from the expected boost pressure map, what happens? => Protection function activated for any deviation? (Error 575: Intake manifold pressure difference) => Protection function only activated when a maximum value is exceeded? I don't know, it can vary depending on the software. There's probably an error entry, and a limp-home mode can also be triggered by insufficient boost pressure, but that was an AFN engine with a VTG turbocharger. The easiest way to try it out

.

Your explanations help me a lot, thank you.

It seems that actually trying it out will probably be necessary if you want to know. Audi probably won't reveal the details of the software version of the AAT, even if asked.

I seem to recall reading somewhere that the 1Z engine can be operated without a boost pressure control system (with boost pressure limited to 0.6 bar by an internal wastegate control) without the engine control unit (ECU) going into limp mode. However, I'm not entirely sure about that.

My AFN immediately goes into limp mode if the turbocharger boost pressure deviates from the specified map; I know this from my own experience with an AFN engine.


Regards,
Jens.

Quote:

Malte1408 wrote the following:: [Please provide the text you would like me to translate from German to English.] "Zitat:" translates to "Quote:" in English. If the control unit detects significant deviations from the expected boost pressure map, what happens? => Protection function activated for any deviation? (Error 575: Intake manifold pressure difference) => Protection function only activated when a maximum value is exceeded? I don't know, it can vary depending on the software. There's probably an error entry, and a limp-home mode can also be triggered by insufficient boost pressure, but that was an AFN engine with a VTG turbocharger. The easiest way to try it out. Your explanations are very helpful, thank you.

I already wrote that a few posts above, before I strayed off topic and went into too much detail.

Quote:

With lower boost pressure (i.e., below the specified range), a limp-home mode can also occur, but it depends on the engine and control unit. In my case, there is no limp-home mode if the boost pressure is too low due to a disruption in the electrical connection to the N75 valve.

I'm going to try to write less here. The basics of how a TDI engine works should be well-known by now, but apparently that's not always the case, or maybe people just aren't interested anymore...

Malte1408 wrote:

Quote: With lower boost pressure (i.e., below the specified range), a limp-home mode can also occur, but it depends on the engine and control unit. In my case, there is no limp-home mode if the boost pressure is too low due to a disruption in the electrical connection to the N75 valve. I'm going to try to write less here. The basics of how a TDI works should, in theory, be well-known by now, but apparently that's not always the case, or maybe nobody is interested anymore...

Nana, now you're being praised, and that's not right .
I understand how a TDI engine works in principle. However, I haven't been able to figure out exactly what the different mapping tables are and what they control and regulate. So, I think it's great that you explained it.

I probably read that your TDI doesn't go into limp mode due to low boost pressure. Since it belongs to a later generation (later than the 1Z and AAT), I didn't consider that information relevant to my situation.

Thank you and best regards.
Jens.

Quote:

Nana, you're already being praised, and that's not right icon_wink.gif

I'm sorry, I seem to have misunderstood something. I probably interpreted the second-to-last post with the quote somewhat ironically. It's as if the only sentence in what I wrote that was worth anything was the one that said I know nothing.

That comment about the basics wasn't really directed at you.

Quote:

I probably read that your TDI doesn't go into limp mode with too low boost pressure, and since it belongs to a later generation (later than 1Z and AAT), I didn't take that information to heart.

Okay, back to the topic. I only know that mine doesn't go into limp mode when the N75 valve is electrically disconnected. I think it's possible that it "knows" that, with the N75 valve electrically disconnected, it might have too little boost pressure, and therefore the limp mode is disabled (as long as the boost pressure is actually lower). After all, the N75 valve error is already registered.

Therefore, disconnecting it might be a way to prevent the emergency mode in your thought experiment, as long as the pressure remains around 1500 millibars.

The next step to verify this would have been to connect a resistor or something similar to the connector while the N75 valve was disconnected, so that it could detect any broken wires.
Then you could test how long it can operate with a pressure of only 400 mbar, which is significantly below the rated pressure, under full load.
However, the effort wasn't worth the realization I gained.

Regards,
Malte.

A small note from me:
At approximately 0.6 bar, the boost pressure in our vehicles is already completely unregulated (the wastegate doesn't activate until 0.6 bar). Therefore, whether the boost pressure control system is disconnected or not makes absolutely no difference and will definitely not cause a limp mode.

The control system doesn't become active until the pressure exceeds 0.6 bar. The standard pressure setting (LD) is 0.9 bar. If your mechanical control system only roughly corresponds to the boost pressure control system, you shouldn't experience a limp-home mode. The pressure deviation at which a limp-home mode is triggered is already relatively wide in our systems.

I don't have the exact deviation in mind, but I seem to recall something about 200 mbar for more than 6 seconds, but that's just a rough estimate and not an absolute number. Your mechanically controlled LD (likely referring to a load dump) should also be regulated to 0.9 bar at full load, so you shouldn't have any problems there either. Anything below 0.6 bar is also not ideal. That leaves only the narrow range of 0.6-0.9 bar. And I find it questionable that your mechanical regulation would deviate so far from the target for such a long time. (It essentially boils down to either the MSG [likely referring to a monitoring system] requiring 0.9 bar, but your mechanics only produce less than 0.7 bar, or the MSG requiring 0.6 bar, but your mechanics produce 0.9 bar).

Therefore, my assumption is that it should also work mechanically without requiring constant operation. However, your test will provide the definitive answer.

Regards,
Guste.