Serial Loader in ASV: Pump Limit Explained

Hello,

I've made an interesting observation (or maybe?) about the boost pressure behavior of my ASV.
The turbocharger is standard equipment and is typically referred to as a VNT 15. (Garrett may also use a VNT 17xxxx model designation.)

I have a Speed-Buster box (I/S/T) and the ability to increase the boost pressure using a diode trick.

When I drive with the standard boost pressure, but have my engine configured to inject additional fuel at full throttle starting from 1800 rpm, there's a noticeable shudder in the intake manifold. I've always assumed this was due to the overstressed dual-mass flywheel (DMF), but if you look closely at the boost gauge, you'll notice that the boost pressure fluctuates significantly at that moment (my gauge doesn't usually fluctuate like that).
Therefore, I assume that at this operating point, exactly what is referred to as "pumping" occurs.

These pumps can be adjusted by moving the threshold at which additional fuel injection is released, either forward or backward.

I now have the impression that perhaps the much-praised ZMS (torque management system) is not the cause of this jerking in the powertrain, but rather the turbocharger, which is being operated at its performance limit.

Perhaps the term "pump limit" isn't clearly defined here, which might lead some people to not understand what I mean by it.
I mean that there are two operating points in the compressor characteristic curve that lie on a horizontal line.
So, a constant pressure ratio, but with two different options for the impeller speed.
If these two points are not separated by a sufficiently high "energy barrier," the charger can oscillate between these two operating states.
During this process, short-term flow separations occur on the impeller blades, and when the flow reattaches, high moments are transmitted.

What do you think about that?
Best regards, Georg.

Georg_G wrote:

This pumping can be controlled by adjusting the threshold at which additional fuel injection is released, either moving it forward or backward.

An interesting observation.
Does that mean that if you release the extra tuning power at 1700 or 1900 rpm (instead of 1800), the "rumbling" or "vibration" will no longer occur? I call it "rumbling."

Quote:

I mean that in the compressor characteristic curve, there are two operating points that lie on a horizontal line. So, a constant pressure ratio, but two different options for the speed of the charger.

As far as I know, that doesn't exist, as long as you don't modify the hardware, i.e., the engine.

The compressor speed increases with both pressure and mass flow rate. However, in the lower range of mass flow, the speed is almost solely dependent on the pressure.

Quote:

Short-term flow separations occur on the impeller blades, and when the flow reattaches, high moments are transmitted.

This is the description for the pumps.

The process:
1. With a low mass flow rate and a high charging pressure, the flow separates (the compressor "surges").
2. The pressure drops (due to air leakage through the engine or even back through the compressor).
3. If the pressure is sufficiently low, the compressor regains its "grip" and builds up pressure again until the flow breaks down again -> No. 1.

It's also possible that your device isn't quite sure whether it should inject more or significantly more.

You should measure the pump voltage in that specific situation; it might be oscillating, and that could be causing the boost pressure to fluctuate.

When running at full load, the EDC creates a plateau (specifically, the torque limit). If the engine now "fidgets" around on that plateau, your Ibiza will start to vibrate.

What kind of performance are you currently offering?

The colleague's ASV is running smoothly with a Box, producing a measured 143 horsepower (without modifying the turbo).

m;

Hi,
In such cases, moments are only transferred from the turbine to the compressor wheel. Thanks to the shaft bearings, the rest of the engine is not affected by this. Furthermore, the fluctuating axial pressure on the turbine bearings can be considered negligible compared to the forces generated by the drive.
It is more likely that the LMM (likely mass flow meter) reacts to fluctuations in the flow rate, which in turn causes the EDC (likely electronic control unit) to initiate nonsensical control actions.

Die Pumpgrenze des GT1749VA liegt bei ca. 2.2bar Ladedruck (getestet). Hier fängt die Nadel der Ladedruckanzeige trotz eingebauter Resonanzdrossel an zu flattern.. In diesem Betriebspunkt reißt, wie schon beschrieben, der Luftmassenstrom ab und wird in den Ansaugtrakt zurückgedrückt. Dieses geht immer solange, bis der Ladedruck für den Lader reproduzierbar wird und er dann wieder die Pumpgrenze erreicht (GT1749VA ca.2,2bar). Dann geht das Spiel von vorn los usw... Dieser Betriebszustand ist allerdings wegen der schon weit überschrittenen zulässigen Höchstdrehzahl besonders kritisch. Wellenschwinger und 'wachsende' Turbinenäder führen dann zu Laderausfällen. Problematisch ist ebenfalls der entstehende Abgasgegendruck. Diesen Betreibszustand erreicht der Lader bloß durch Nutzung der max. Abgasmenge vom mittleren Drehzahlsbereich des Motors an ohne Regelung durch VTG oder Wastegate (Regelorgane für konstruktive Unterdimensionierung der Lader für gutes Instationärverhalten). Da die Serienlader nicht ausreichend axial gelagert sind, verstärkt der Pumpeffekt durch fehlenden Ansaugegendruck die axiale Kraftwirkung in Richtung Turbine. Die Folge ist ebenfalls ein Laderschaden,der sich an Hand eingedrückter Verdichtergehäuserückwände zeigt. Vorsicht! Die The risk of charger damage due to compressed air being forced back into the charger, caused by increased exhaust back pressure, already exists with excessive tuning modifications (e.g., pump, injectors, etc.).

That kind of waste is not what mass-produced products are designed for, and yet it's still expensive .

Best regards!

christians wrote:

Hi, Moments are then, at best, transferred from the turbines to the compressor wheel. Thanks to the roller bearings, the rest of the engine doesn't feel any of it.

Hello Christian,

I also meant the torque on the shaft that connects the turbine and the compressor. That one is not as thick, and that's what I'm worried about .


@ DieselMartin
I can rule out oscillating additional fuel injection.
I can also reproduce this behavior with a 10cc fuel injection tuning adjustment.

143 horsepower in the ASV without increasing boost pressure? Doesn't that seem a bit low?
My performance is roughly equivalent to Pi*Thumb, with acceleration times in 3rd gear well below 5 seconds.

@Ulf
Yes, that knocking sound disappears if you allow for a higher fuel injection amount earlier on (at a lower boost pressure).
This "pumping" phenomenon is characterized by (low flow on the turbine side) combined with (a high pressure ratio on the compressor side).
If you increase the fuel injection amount, you will also see more flow on the exhaust side.

@Dieselrudi:
What do you mean by 2.2 bar? Is it absolute pressure or relative to atmospheric pressure?


Best regards, Georg.

Of course, it's overpressure. Otherwise, with the atmosphere, it would be the standard loading pressure of some PDs.


Best regards!

Hi Georg,

To achieve a sub-5 second 0-100 km/h time, you'll need more than 143 horsepower. My colleague's Ibiza, as far as I know, had a time closer to 6 seconds (but with 3 people and a slight uphill incline).

The exhaust sound is relatively quiet - yes, you can tell when you're behind him and he's constantly accelerating. But stock PD and CDI models tend to make more noise when starting off gently.

m;

Hello,

'That's interesting! Currently, I'm using a 12mm fuel pump with AFN injectors and a VNT15 turbocharger. I don't have time for further modifications to the turbo right now, so I've already reduced the fuel quantity to the point where it only produces black smoke in two specific situations, due to the large fuel flow of the pump.'
Once, at full throttle in 5th gear, but only very slightly. That is acceptable.
And once, when accelerating with full throttle in the 4th or 5th gear at 1900-2200 RPM.
'I also experience this 'roughness' and previously thought it was the ZMS (torque converter). And indeed, the turbocharger isn't building boost properly at that moment. It feels like turbo lag, and the car emits a thick black cloud from the exhaust before it starts moving forward again. I haven't installed a DPF (diesel particulate filter). The same procedure in 2nd or 3rd gear doesn't produce any black smoke. If I give it full throttle in 4th gear at 1400-1500 RPM, it doesn't happen either. So, it only occurs under high load around 2000 RPM.'
The charger is definitely operating right at its limit. It's experiencing overvoltages up to 1.6 bar.
'To be honest, I don't really understand what the pump limit actually means in a practical sense. Would using a larger pump solve the problem?'
The maximum limit is probably already reached (keyword: speed of sound). The car sounds like a jet engine above 3500 RPM . The VNT20 must come, no question. It would be good if this phenomenon then disappeared.
I would appreciate an explanation regarding the pump limit.

Regards,

Maciek.

Hello,

I've re-read the entire thread, and the pump limit seems clearer now. So, with a VNT15 impeller, I need more engine speed to achieve a certain boost pressure compared to a VNT20 impeller. If the desired boost pressure is too high (outside the VNT15's specifications), it will cause such a high impeller speed that the airflow separates, and the turbocharger overspools. Then, it loses speed, 'recovers,' and the boost pressure builds up normally again, until it potentially reaches the pump limit again. Therefore, a larger impeller should eliminate this problem. With my VNT15, the problem might disappear if I make the VTG adjustment 2000 RPM less aggressive. The current response is probably too abrupt.
Does that make sense?

Regards,

Maciek.

Maciek wrote:

If the desired boost pressure is too high (outside the specification of the VNT15), it can lead to such a high RPM that the airflow separates, and the turbocharger spins freely. Then, it loses RPM again, "recovers," and the boost pressure builds up normally again, until it may reach the pump's limit again. ... In my case with the VNT15, it might disappear if I make the VTG adjustment 2000 RPM less aggressive. The response is probably too sharp right now. Is that correct?

Based on my experience with the AFN + VNT15, it's definitely worth trying to machine the VTG rod to a longer length.

Maciek wrote:

Hello, I've reread the entire thread, and the pump limit seems clearer now. So, with a VNT15 compressor wheel, I need more RPM to achieve a certain boost pressure compared to a VNT20 compressor wheel. If the desired boost pressure is too high (outside the specification of the VNT15), it will lead to such a high RPM that the airflow separates and the turbocharger overspools.

Hello Maciek,

That's roughly what I imagine as well. However, in my opinion, flow separation only occurs when there is insufficient flow in the turbine. This is typical for low speeds and simultaneously high boost pressure, as is required for our TDIs at 2000 rpm.
I would like to discuss this using a compressor performance map/diagram. I will select one and add some annotations to it.
I'm still not entirely sure if Ulf, Christian, DieselMartin, etc., when they talk about "Pumpgrenze," are referring to the same thing as what is understood by the English technical term "compressor surge." I understand it as two operating points in the compressor characteristic curve that lie at the same pressure level (like an isobar in a pV diagram), but with different turbine speeds. Between these two operating points, the compressor oscillates back and forth, meaning it's trying to achieve the same pressure ratio, but with one speed and then the other. I think I'll try to draw a diagram.

My understanding of the problem (insufficient flow + high pressure generated by the compressor wheel) is supported by the fact that when I release the additional fuel injection at around 1600 rpm, the knocking noise decreases. This is because the increased exhaust flow from the turbine then becomes more significant.

P.S.
Does anyone actually know the exact difference between the ASV turbocharger GT1749V and the 130 HP common rail turbocharger GT1749VA? As far as one can tell from the pictures on eBay, the compressor wheels look the same.

Best regards, Georg.

Hello Georg,

I mean, all this mess happens like this :
The compressor wheel is driven to such a high speed by the 'aggressive' positioning of the guide vanes on the diffuser side of the VTG and an excessive injection rate that it 'empties' the intake side. There's a physical limit: with a given inlet diameter in the compressor housing and a specific compressor wheel design, a maximum amount of air can pass through in a given time. If you reach this limit, you can rev the turbine as much as you want, but the airflow will detach, and you'll be creating a 'vacuum.' Simply not enough air is coming in, and the compressed air flows back into the 'vacuum' until you're below that limit again. This would likely cause severe turbulence in the compressor wheel.
When running at full throttle from 2000 RPM, the combination of VTG (Variable Turbine Geometry) position and the allowed fuel injection amount will reach the pump's limit. If starting at 1500 RPM, the allowed fuel injection amount is likely still too small to accelerate the compressor wheel to the pump's limit, and by the time 2000 RPM is reached, the turbocharger system will have already adjusted (reduced boost).
That's just my layman's understanding of it.
That concept of having two operating points with the same boost pressure but different speeds doesn't quite fit with my understanding. But maybe you can find a plausible explanation.

Regards,

Maciek.

Hi,

Given that the intake manifolds have a considerable cross-section, the inlet before the air filter is the bottleneck, and only 50 mbar or less of negative pressure has been measured in the air filter housing so far, I don't believe that any significant vacuum can be created there.

Best regards,

Jan.

Hello Jan,

Vacuum in ' ' . I don't imagine it happening like this, where the air housing is suddenly compressed. These are also only fractions of a second. The air is accelerated so strongly at the compressor inlet that the flow separates and the supply stops. This is a very localized 'vacuum' . You can see on a compressor diagram that, for the same boost pressure, a lower air mass requires a higher speed than a higher air mass (if I'm interpreting this correctly). The 'vacuum' is a critical air mass limit below which even a higher boost speed won't help. I now firmly believe this .

Regards,

Maciek.