Exhaust Gas Recirculation (EGR) flow direction in case of malfunction

Hello!

As previously discussed internally with ulf, I would like to refute the theory regarding the flow direction within the EGR system in the event of a malfunction.

It is said that when the EGR valve is open under full load or high load, a portion of the fresh air bypasses the combustion chamber and goes into the exhaust system, thus not participating in the combustion process.

For several reasons, this theory must be incorrect.

- The boost pressure cannot be higher than the back pressure in the exhaust system before the turbine, because otherwise the engine would run even without injected fuel, or the turbocharger would act as a perpetual motion machine if you connect the fresh air outlet to the exhaust inlet.
A turbofan engine always exhibits a significant negative spillage gradient.

- During acceleration, the actual air mass deviates significantly upwards from the target air mass, despite the open EGR valve. Why? Because the pressure upstream of the turbine is too low to force enough exhaust gas through the EGR system into the intake manifold, or because the pressure difference between the exhaust manifold and the intake manifold is not sufficient. Once fuel injection resumes, the actual air mass is correct again.

- For those who still have doubts, please refer to the air mass flow curve and the torque limitation characteristic in the data set of a 1.9 TDI AVF engine with a 019 LJ control unit. The torque limitation characteristic has a maximum value of 56 mg/cycle, but the air mass flow curve only reaches 850 mg of fresh air/cycle starting from 51 mg/cycle.
Here, the AGR (Abgasrückführung - Exhaust Gas Recirculation) remains open almost to its maximum limit, and within the range of 2600 to 3432 RPM, there is an overlap where the AGR is even fully open at full load.

"If the generally accepted AGR theory were correct, the AGR pipe would be oily on the inside. However, if there is a loss of boost pressure (in the event of a fault, or a deliberately induced fault) and it passes through the AGR pipe, it will instead be dry and completely black, which contradicts the theory."


Okay, and now I'm curious to see what you all think...

Regards,
John.

vwSchrauber wrote:

The boost pressure cannot be higher than the exhaust backpressure before the turbine, because otherwise the engine would run even without injected fuel, or the turbocharger would act as a perpetual motion machine if you connect the fresh air outlet to the exhaust inlet. A turbomotor always exhibits a very pronounced negative scavenging gradient.

Okay, I'm going to present my thoughts here, arguing against the idea that there is a flow direction reversal (at least) at full load in an EGR-FA system.
This discussion was actually coordinated with someone familiar with Volkswagen engines and is intended to simply provide clarity and, if necessary, lead to a modification of the EGR valve.

The explanation regarding the drain slope is missing the thermal aspect, specifically the significantly higher volume flow of the exhaust gases that are expanding compared to the compressor side.
The "side stream" performance of a turbine is calculated, as far as I know, as the volumetric flow rate multiplied by the pressure difference between the inlet and outlet.
Given the massive volume flow of exhaust gases, a relatively small pressure difference above the turbine should be sufficient. Therefore, it seems to me that it is definitely possible to generate more lift (LD) than the (average - due to pulsations) backpressure of the exhaust gas.

Quote:

- Under full load, the mass airflow sensor deviates significantly upwards from the target air mass, even with the EGR valve open. Why? Because the pressure upstream of the turbine is too low to force enough gas through the EGR system into the intake manifold, or because the pressure difference between the exhaust manifold and the intake manifold is not sufficient. As soon as fuel injection resumes, the air mass reading becomes correct again.

Objection: The target air mass is arbitrarily set by the programmer and therefore not a reference for possible or impossible operating conditions.

Quote:

- Those who still have doubts should please examine the air mass target map and the torque limitation curve in the data set of a 1.9 TDI AVF engine with a 019 LJ control unit. The torque limitation curve has a maximum value of 56 mg/cycle, but the air mass map only reaches 850 mg of fresh air/cycle starting from 51 mg/cycle. Here, the AGR remains open almost to its maximum value.

This also gives me something to think about.

Quote:

In theory, the engine should produce a lot of soot when running close to full load - unless its lambda control is adjusted for fresh air losses, or simply depends on the load instead of the measured airflow (in the EDC15, there's a software switch for this).

- If the generally accepted EGR theory were correct, then the inside of the EGR pipe should be oily if there is actually a loss of boost pressure (in case of an error, or a deliberately induced error). However, if it is dry and completely black, this contradicts the theory. In practice, this will hardly lead to clear results, because under partial load, the EGR system will certainly operate correctly and deposit soot into the pipe. To create a clearly visibleoil film in the pipe (with reversed flow direction), the engine would have to run for a very long time} at high load with the EGR valve fully open...

Quote:

In practice, this is unlikely to lead to clear results, because at partial load, the EGR system will certainly operate correctly and carry soot into the pipe. To create a clearly recognizable oil buildup in the pipe (with reversed flow direction), the engine would have to run for a very long time under high load with the EGR valve fully open...

Here's a solution: Connect the EGR valve to vacuum. This will keep the EGR valve fully open all the time. If there's ever too much pressure in the intake manifold, it will definitely flow in the wrong direction.

vwSchrauber wrote:

Connect the EGR valve to negative pressure, and then the EGR will always be fully open. If the pressure in the intake manifold is higher, it will definitely flow in the wrong direction.

Yes, but how would a TDI engine possibly run with its EGR permanently open?
Based on experience with similar defects, it will hardly pull at all, but it will produce a lot of black smoke *).

*) Note: The asterisk suggests a footnote or further explanation might be needed for "rußen" (to soot/produce soot).
I certainly wouldn't want to push it so hard for that long, until a clear oil film has formed from the KGE (Knock Gas Emission) vapor in the EGR (Exhaust Gas Recirculation) pipe. My gut feeling is that it would take at least half an hour...

*) Within this thread, a plausible explanation for why this is happening should also be provided. The most obvious approach, in my opinion, is the reversal of airflow, which acts like a boost pressure leak in terms of fresh air loss.
All the fresh air drawn in is reported by the mass airflow sensor (MAF) as fuel, but less of it actually reaches the cylinders than is needed for a soot-free combustion of the full load amount.
Without exhaust gas recirculation reversal (i.e., your theory), the LMM signal would be reduced, and the engine would likely enter a state where it's prone to excessive soot production – which is precisely what the EGR system is designed to prevent. So why do TDIs with stuck-open EGR valves still produce so much soot?

Quote:

I estimate, just based on a gut feeling, that it would take at least half an hour.

I think it could also handle some strong engine loads. If that much air is being lost through the EGR system, it would definitely pull enough oil along with it.

Quote:

All the fresh air that is drawn in is reported by the mass airflow sensor (MAF) as combustion material, but less of it actually reaches the cylinders than is needed for a soot-free combustion of the full load amount.

Firstly, local areas of over-saturation can occur due to a lower density of oxygen molecules.
On the other hand, even with the maximum air mass measured by the mass airflow sensor (MAF), it's not necessarily the end of the story. This is particularly noticeable in TDIs with high displacement, but even the older 1Z engine, purely based on calculations, achieves around 2000 1/min more than 850mg/cycle under full load.

vwSchrauber wrote:

I mean that a few strong engine loads would also do it. In that case, it would already pull enough oil up if really that much air is lost through the EGR valve.

Hmm, extrapolated to the oil consumption based on mileage, this description makes me think of a real oil guzzler.
I can't imagine that just a few strong bursts of pressure would draw so much oil from the wetted intake passages into the EGR pipe that it becomes noticeably oily.

Quote:

Firstly, local over-enrichment due to a lower density of oxygen molecules.

. . . which, however, should result in a "suitable" reduction of the injection amount via LuMa turbidity...

Quote:

On the other hand, even with the maximum air mass measured by the MAF sensor, it's not necessarily the end. This is more pronounced in TDIs with high displacement, but even the older 1Z engine theoretically produces around 2000 1/min more than 850 mg/cycle under full load.

I am aware of that, but the diagnostic scale only goes up to 850 mg.
But what are you trying to get at with this comment?

vwSchrauber wrote:

Bitte gib mir den deutschen Text, den du übersetzt haben möchtest. For several reasons, this theory must be incorrect. - der Ladedruck kann nicht höher sein als der Abgasgegendruck vor Turbine, da sonst der Motor auch ohne eingespritzten Kraftstoff laufen würde,...

Hi,

Sorry, you're mistaken. If that were the case, not a single turbofan engine would work.
There was this series about a group of mechanics who built an engine using a large truck turbocharger.

Regards,

Edit:

I found out how it works: http://de.wikipedia.org/wiki/Gasturbine

Quote:

Hmm, extrapolated to the oil consumption for running performance, this description makes me think of a terribly wasteful oil guzzler. I can't imagine that just a few strong engine loads would draw so much oil from the wetted intake passages into the EGR pipe that it becomes noticeably oily.

Considering that when an intercooler hose bursts, everything around it is covered in oil, the leak can't be very small.
The oil consumption might seem surprisingly low because diesel can be expelled from the engine during operation, which had previously entered through the blow-by system during a cold start. This is especially true during direct injection (DI).
After the intercooler, the fuel that has vaporized condenses again in the intake manifold.

Quote:

. . which, however, should result in a "suitable" reduction of the injection quantity via LuMa blurring . . .

Quote:

I am aware of that; only the diagnostic scale ends at 850 mg. Aber worauf willst Du mit diesem Hinweis hinaus?

The scale ends at 850mg. However, the TDI (depending on the version) requires more than 850mg under full load to burn cleanly.
As you mentioned, the diagnostic scale ends at 850mg, but since this value is reached even before full boost pressure is achieved, the air mass would continue to increase (which also happens in newer TDI engines).
Looking at the reading of 850mg on the tester, one might be tempted to believe that the engine is receiving enough fresh air, but this may not necessarily be the case.

Local over-fueling: This is like injecting fuel into combustion chamber areas where combustion has already occurred, and all the oxygen molecules present there have been consumed. In production engines, this is largely prevented by the swirl (mainly caused by the swirling motion).
However, if there are too many inert gases hindering the rapid formation of the air-fuel mixture, it can lead to the production of black smoke.

Quote:

sorry, da hast einen Denkfehler. Wenn das so wäre, würde kein einziges Turbinentriebwerk funktionieren.

Considering this aspect, you are absolutely right.

Take a look at the characteristic curves for pressure ratios and efficiency in a turbocharger. You'll see that as the gas flow increases, the backpressure required to maintain the desired boost pressure also increases.

You can really feel the difference in how smoothly they turn over, especially when comparing the AGR and ALH engines in a Golf 4. Both are 66kW TDIs. One has a relatively small wastegate turbocharger, while the other has a comparatively larger variable geometry turbocharger (VTG). "Large" and "small" here refer to the size of the turbine.

vwSchrauber wrote:

The oil consumption can also be deceptively low because diesel fuel is being expelled from the engine, which had previously entered through the blow-by during a cold start. This is especially true in direct injection (DI) engines. After the coolant heat exchanger (LLK), the evaporated fuel condenses again.

. . . but if it does, then without the components that create the typical diesel smell...

Quote:

Local over-fueling: This is like injecting fuel into combustion chamber areas where combustion has already occurred, and all the oxygen molecules present there have been consumed. The swirl (mainly due to the swirl motion) largely prevents this in production engines. However, if there are too many inert gases hindering the rapid mixing of the fuel-air mixture, it can lead to black smoke formation.

Hmm, it seems theoretically plausible... but the (IIRC, although those threads were a long time ago) significant power loss caused by a severely stuck EGR valve suggests "huge" amounts of unburned diesel. How high could that percentage be, according to your theory?
By the way, I checked the 019LJ software: The sootiness seems to be controlled via LMM (mass airflow measurement), and the sootiness control parameters also appear normal, as they would be in engines with "normally early closing EGR". The fact that no anomalies are detectable here seems to support your assumption...

Yes - but how would a TDI engine likely run with the EGR valve permanently fully open?


My V6 engine was running terribly with that fuel. It had severe misfires, and anything above idle speed resulted in white smoke.

Quote:

A significant performance loss with a stuck-open EGR valve indicates "huge" amounts of unburned diesel fuel. According to your theory, how high could this percentage be?

Quote:

und alles über Standgas wurde in Weissen Nebel verwandelt.

Given the white smoke, there is likely insufficient oxygen available to reach the combustion temperatures necessary for complete ignition. For black smoke, at least in the area of the jet streams, the temperature must be high enough to initiate combustion (which would then be incomplete).

The proportion of unburned diesel is difficult to estimate, perhaps around 30%?
It should be noted that diesel fuel that has only evaporated (producing white smoke) removes heat from the combustion process and therefore acts as a "negative fuel."

Quote:

By the way, I checked the 019LJ software: The soot seems to be flowing via the LMM, and the soot control valve also looks completely normal, like in engines with "normally early closing EGR". The fact that no anomalies are detectable here seems to support your assumption.

Not the smoke map. My theory is based on the air mass flow map, which specifies the maximum amount of air that can enter the combustion chambers.

vwSchrauber wrote:

Nicht das Rauchkennfeld. Meine Theorie stützt sich auf das Luftmassenkennfeld, welches vorgibt welche Luftmasse max. in die Brennräume darf.

. . . das AGR-KF, ist mir schon klar. Aber meinte die Ruß-Erklärung über Ladeluftverluste via AGR, siehe oben:

Quote:

eigentlich müßte der Motor knapp unter Volllast stark rußen - es sei denn, sein Trübungs-KF wäre auf die Frischluftverluste angepaßt

. Aber das It looks normal, as mentioned before.

An alternative testing method is swirling around in my mind.
Log the LD control (MWB11) at full load, as well as the lambda value, opacity, emissions, and pedal position. Then, using a temporary button, briefly open the EGR valve completely and check the log to see what happens: according to my theory, the lambda value should increase, while your theory suggests it should decrease.

I'm just hesitant to cut into my wiring harness to integrate the button (and avoid error messages from the electrical monitoring of the EGR valve).

For the sake of completeness: At that time, I had used a 1N4148 diode instead of the LMM (likely referring to a specific component).
The company permanently opened the EGR valve to its maximum extent. Therefore, in at least some cases with V6 engines, the diode conversion was always accompanied by deactivation of the EGR system.
With a functioning lambda sensor (LMM), the white smoke would likely be present to a similar extent.

vwSchrauber wrote:

The scale ends at 850mg. However, the TDI (depending on the version) requires more than 850mg under full load to burn cleanly.

vwSchrauber wrote:

Der Ölverbrauch kann auch deswegen unauffällig gering sein, da Diesel aus dem Möl rausgekocht wird, der zuvor übers blow-by beim Kaltstart eingetragen wurde. Gerade am DI ist das nicht mal so wenig.

These are two theories that I would reconsider.
A diesel engine requires a minimum amount of fuel to combust properly; otherwise, features like pilot injection and idle operation would not be possible.
I doubt that unburned diesel fuel, in the form of liquid diesel and not soot, would be able to pass from the combustion chamber, past the piston rings, and into the engine oil.
The temperatures, even with incomplete combustion, are simply too high for diesel fuel to remain in a liquid state.

But let's get back to the topic.
My experience has shown that a non-closing AGR valve at full throttle and higher RPMs is often associated with excessive carbon buildup. This can only happen if the MAF sensor (which has not been replaced with a diode) reports sufficient fresh air intake.
But where should the gases go if not past the combustion chamber (on the outside) and into the exhaust?
One should also not disregard the principles of fluid dynamics, which state that air creates a vacuum with increasing flow velocity. In this case, the airflow at the EGR valve will be higher on the exhaust side than on the intake side.
Furthermore, the turbine wheel is designed with larger dimensions, which allows for a higher pressure to be generated on the compressor side even with lower back pressure.

Quote:

An alternative testing method is swirling around in my head: Log the LD control (MWB11) at full load, as well as the lambda value, opacity, emissions, and pedal position. Then, using a temporary button, briefly open the EGR valve completely and check the log to see what happens: according to my theory, the lambda value should increase, but you think it might decrease. Ich sträube mich nur, meinen Kabelbaum anzuschnippeln, um den Taster einzuschleifen (und Fehlermeldungen aus der elektrischen Überwachung des AGR-Ventils zu vermeiden) . . .

Insert a T-connector into the pneumatic line leading to the valve manifold and attach a hose to it.
Connect the other end of the hose to the vacuum line (there are usually a few unused ports on the check valve). An assistant in the passenger seat holds the loop of the hose (over the window) and keeps it bent. Under the right conditions, they straighten the bend, allowing the vacuum to reach the valve body.
A suitable material for this would be a plastic hose, similar to those used in various pneumatic logic systems. I'm familiar with products from Festo that have the designation "PUNxx," where "xx" indicates the diameter. These hoses can be easily bent completely and then straightened again, and they can withstand repeated bending.

Tagessuppe wrote:

These are 2 theories that I would think about again. "Does a diesel engine require a minimum amount of fuel to burn cleanly? Otherwise, things like pilot injection and idling wouldn't be possible."

The 850mg refers to the mass of air being drawn in. If you were to put 850mg of diesel into a 1.9 TDI engine, the sun wouldn't rise the next day .

Quote:

2. I doubt that unburned diesel fuel, in the form of diesel and not soot, would be able to pass from the combustion chamber, past the piston rings, and into the engine oil. The temperatures, even with incomplete combustion, are simply too high for diesel fuel to remain in a liquid state.

"During cold starts and the subsequent warm-up phase, this phenomenon definitely occurs. Therefore, engine developers are careful to program an appropriate afterglow phase."

2.) Modern diesel engines with particulate filters sometimes have significant problems with fuel entering the engine oil, which can even lead to an increase in the amount of engine oil.
This occurs because the fuel injected relatively late in the regeneration process, which is necessary for that process, only partially combusts. The remaining portion of the fuel comes into contact with the cylinder wall and seeps into the engine oil through the piston rings.

Even diesel engines without a DPF can produce varying amounts of soot, depending on the injection timing and the condition of the fuel injection system (spray pattern).

Quote:

In this case, the flow at the EGR valve will be higher on the exhaust side than on the intake side.

And how do you explain the theory for those DIS systems where the AGR pipe runs next to the exhaust outlet of the 4th cylinder? For example, the Audi A4 with a (longitudinal) 1Z engine.

Quote:

Ausserdem ist das Turbinenrad grösser Dimensioniert, weshalb mit geringerem Staudruck ein höherer Druck an der Verdichterseite erzeugt werden kann.

Only applies to VTG (Variable Turbine Geometry) chargers. In the Garrett T15 and GT15 turbochargers used in various wastegate-equipped TDIs, the exhaust side is smaller, and the EGR (Exhaust Gas Recirculation) issue functions in the same way.

vwSchrauber wrote:

Quote: 2. I doubt that unburned diesel fuel, in the form of diesel and not soot, would be able to pass from the combustion chamber, past the piston rings, and into the engine oil. The temperatures, even with incomplete combustion, are simply too high for diesel fuel to remain in a liquid state. "During cold starts and the subsequent warm-up phase, this phenomenon definitely occurs. Therefore, engine developers are careful to program an appropriate afterglow phase."

Do you really believe that a tiny glow plug has more influence on verschandeln combustion than a proper air excess?
I highly doubt that. In my AKE (presumably a type of device), the afterglow lasts for a maximum of 5 seconds.
and I already think that's an exaggeration.

Quote:

2.) Modern diesel engines with particulate filters sometimes have significant problems with fuel entering the engine oil, which can even lead to an increase in the amount of engine oil. This occurs because the fuel injected relatively late in the regeneration process, which is necessary for that process, only partially combusts. The remaining portion of the fuel comes into contact with the cylinder wall and seeps into the engine oil through the piston rings. Even diesel engines without a DPF can produce varying amounts of soot, depending on the injection timing and the condition of the fuel injection system (spray pattern).

Do you have any reports that support this or a theory explaining why it specifically has to do with the DPF?

Quote:

And how do you explain the theory for those DIS systems where the AGR pipe runs next to the exhaust of the 4th cylinder? For example, the Audi A4 with the (longitudinal) 1Z engine.

It's just an additional factor that I can only report from my own experience with a small AGR motor.
There, I personally experienced what happens when the fresh air intake is blocked due to a malfunctioning EGR valve.
So, you would be the one to disprove that circumstance.

Quote:

Nur an VTG- Ladern. Beim Garrett T15 bzw. GT15 von diversen Wastegate- TDIs ist die Abgasseite kleiner, und das AGR- Problem funktioniert da genauso.

Regarding my wastegate turbocharger on the AGR TDI, the exhaust side is significantly smaller compared to the VTG turbocharger, but still larger than the compressor side.
One shouldn't rely solely on the outward appearance in this case.
Because VTG turbochargers are primarily larger on the exhaust side because the VTG adjustment mechanism is located externally around the turbine wheel in that area.
Furthermore, it's not the overall external size that matters, but rather the size of the turbine wheel and the compressor, and even then, it's not just the outer diameter, but the total surface area of the blades that needs to provide the driving force.