Daten zum 2.0 16V mit Filter | Posts 32+

Quote:

It is not possible to explain the IMA solely based on the exhaust volume resulting from a short-term injection.

Yes, that seems to be the explanation.
I zoomed in on the section again and included the minimum values of the torques (008.2..4).

As I now know, the VTG remained active (the flickering of the N75 didn't matter).
And lo, the boost pressure runs parallel to the exhaust flow (which is equivalent to the amount of electrical energy).

m;

Jochen_145 wrote:

I just don't understand the cloud graphic, which should actually be a line.

I think that's because the duration of the signals is not discernible in the graphic. If only a slight twitch with a small amplitude occurs at N75, and this twitch is very short-lived, then nothing will happen at G581 either.

I once looked into this motor in the SSP documentation. The G581 is used together with the LD sensor G31. G581 is also relevant to the military.

Martin's fundamental problem remains. I think twisting the VTG rod will do the trick.

Quote:

I think that's because the duration of the signals is not discernible in the graphic.

But

My graphics are all XY scatter plots with seconds on the X-axis, so the numbers aren't "measurement numbers" (I made sure of that).

m;

A 008-011-043 DZR follows...

Hm, maybe I'm just too dumb .
To me, your cloud graphic looks like x = TV-N75 and y = TV-G581. So, for each data point, there's a corresponding N75 value and a G581 value. Whether the signal lasted 0.2 or 0.5 seconds, I can't see that .

rubber

Which of the approximately 2,351,863 images I have posted are you referring to?

m;

Ah, Rubber, now I understand you:

You mean the N75-G581 cloud.
Yes, it doesn't display the actual values, but only the two percentage values.

But it is precisely HERE that time becomes the crucial factor again:
A straight line, as one might imagine and as I also expected, only exists if the VTG immediately implements the commands of the N75 and the G581 measures it.

Since the vacuum sensor is slow to respond, there are delays in the execution - and suddenly, the straight line is "over."

It was clear that it wouldn't be a straight line, because for example, if the N75 value drops from 85% to 77%, and that still means the VTG is "fully closed," then there can't be a purely linear relationship.

Perhaps it would be possible to try to eliminate a kind of inertial component from the measurable N75-to-G581 values, so that we know that the G581 values always lag behind the N75 commands by x,y seconds...

m;

Yes, that's exactly the cloud graphic I'm talking about... because that's exactly what it was about, and that's what I wanted to express. At the end, I probably only caused confusion .

Jochen_145 wrote:

I just don't understand the cloud graphic, which was actually supposed to be a line

.

dieselmartin wrote:

From my gut feeling, I would have expected a bit more coherence between the two signals... It looks quite "cloudy"

.

And in the graph, one cannot see the duration for which the signal was present, which would have been necessary for evaluation.

It's only possible to eliminate the inertia component by considering the change in the duty cycle at the N75. Because if you add a 10% jump, the pressure sensor will only slowly catch up (i.e., it will be quite sluggish). If the TV price jumps by 30%, things will move faster.

Hi.

I have now come to the following opinion:

Trying to relate N75 and G581 doesn't make sense (mainly because of the inertia in between).

Unlike previous TDIs, when considering boost pressure, the N75 valve should no longer be used, but instead the G581 valve. Therefore, a differential pressure sensor (DZR) log should include group 043. Thanks to CAN on the diagnostic port, a log with groups 008-011-043 should provide a large number of highly detailed measurement values.

If the DZR (Differential Pressure Sensor) reading for the G581 is at 85%, but there is still no boost pressure, it is likely due to a narrowly defined boost limit (-> to protect the DPF) and/or the relatively large GT1749VC turbocharger.

m;

dieselmartin wrote:

Trying to relate N75 and G581 doesn't make sense (mainly because of the inertia in between).

Aron wrote:

Vacuum in the system, temperature at the solenoid valve, slight clogging of the VTG (stiffness), and atmospheric pressure on the membrane should all be factored into the control system.

In addition, there's the exhaust volume before the turbocharger, which I think you mentioned before. Even a 400°C hot exhaust manifold is noticeable in cruise mode. There's already more backpressure here than with a cold exhaust manifold, because ultimately the gas temperature before the turbocharger and the amount of air that passes through determine the volume that drives the turbocharger.

Ultimately, the final actuation of the entire VTG linkage occurs at 85% TV. For me, the slight movements of the position sensor almost suggest that, during the build-up of boost pressure, the N75 is being driven into a range where the actuator is not fully at its limit, but only "near" it, in order to react quickly to the impending boost limitation. It's actually quite strange that the TV is reduced so much during acceleration that the sensor only twitches minimally. It would be a strange coincidence if the MSGR held the VTG at the exact point where the sensor twitches during boost build-up; at least, that's what it looks like if you've spent a lot of time thinking about it.

Here, I'm going to "step on the gas" a little.

It's interesting that the N75 is significantly "more hysteric" than the actual movement of the VTG.

MSTG really seems to be "afraid" of excessively high boost pressure.
The target hasn't been reached yet, and it's already showing significant resistance.

Here's a genuine DZR valve with G581 connection.

And one more.
The first five gears (1 and 2 are just for getting started).

I once looked into the SSP documentation for this engine. The G581 is used together with the LD sensor G31. (..)

Hello Rubber,

'Die deutsche Wirtschaft steht vor großen Herausforderungen. Der Fachkräftemangel, die steigenden Energiekosten und die geopolitische Unsicherheit belasten die Unternehmen. Gleichzeitig bieten sich aber auch neue Chancen, beispielsweise im Bereich der Digitalisierung und der Nachhaltigkeit. Es ist wichtig, dass die Politik die Rahmenbedingungen schafft, die Innovationen fördern und Investitionen in die Zukunft ermöglichen.'

'The German economy faces significant challenges. The shortage of skilled workers, rising energy costs, and geopolitical uncertainty are putting a strain on companies. At the same time, new opportunities are emerging, for example in the areas of digitalization and sustainability. It is important that the government creates the framework conditions that promote innovation and enable investments in the future.'
'The G581 is used together with the LD sensor G31 to capture the status of the LD control.'

Being a rule is not a status.
A functioning or defective system, however, has either one or zero error statuses.
Therefore, the G581 is suitable for error detection, as it was explained to me.

@Martin

If the boost pressure setpoint is adjusted so poorly during dynamic motor mapping (DZM), I don't understand the regulator tuning.

The target value should actually be reached much earlier.

Do you experience this behavior even when it's warmer?

Best regards, Jochen.

Sure, here's the translation:

"First, a quick note:"

If the X-axis displays whole numbers, it's not showing seconds; Excel is misleading me.

@ Jochen

I think the weather makes very little difference.
The first graphic (the one with the orange segment labeled "Russgrenze schalegt zu") was posted on August 24th at 18:24.
We had temperatures of 30 degrees Celsius here: http://wetter.eit.uni-kl.de/anfang/historie.html

Regarding your question – why this cautious tuning compared to the 1.9 engine – as you can see from the previous graphs, this isn't about the stressed mechanics of the dual-mass flywheel, etc. That would be illogical, especially since the smaller engine here is under more stress. No, it's simply the immense fear of boost pressure overshoots, and how significant those are in the larger 1.9 engines in their standard configuration – something we all (should) know.