Poor performance at full throttle.

Hello everyone,

In my AFN, I'm experiencing some strange behavior:

For some time, I had the feeling that when I gently pressed the accelerator pedal, for example after coasting in 'LL', the car initially braked. Only with more gas did the acceleration then suddenly start. This reminded me of lean mixture conditions in a gasoline engine. I then made measurements with VAG-Com and discovered that the LMM value was much too high. My search on the forum led me to the AGR valve. I removed it, cleaned it (now it definitely closes), and disabled it. The effect is now gone.
However, the effect remained that at full throttle from a low RPM, the acceleration is extremely lean. When you then slightly release the gas pedal, the acceleration improves. It's also better if you press the pedal slowly.
Furthermore, it seems that the effect is less pronounced when the engine is cold. I checked the actuator rod of the compressor, and it feels smooth. Of course, I can only check them when they are cold, at least with my fingers.
Unfortunately, I don't have a VAG-Com notebook available at the moment, but perhaps someone here can also figure out a solution.

Greetings
Thomas

Hello,

I was now able to perform a measurement with VAG-Com.
When idling, the air mass flow (Block3, Channel2) should be 285, and the air mass flow (Block3, Channel3) should be 480. Everything else appears normal.
At full load (3000 1/min), the air mass should be =800, and the air mass is actually 780. Additionally, the voltage of the control slider (Block1, Channel3) is also completely outside the tolerance at 3.6V.
Now I'm quite at a loss.
Who can help me?

Greetings
Thomas

meyert wrote:

For some time, I had the feeling that when I gently pressed the accelerator pedal, for example after rolling in low gear, the car initially braked. It was only with a bit more gas that the acceleration suddenly came.

Hi Thomas

IMO, this is objectively impossible: In the (gas pedal)idle position, the throttle cutout is active, meaning the engine generates the maximum braking effect.
Therefore, applying even more engine braking (e.g., when gently pressing the accelerator) is not possible.

Probably, you're expecting acceleration even though nothing has been injected yet (when the train only starts moving again at medium or high speeds when the pedal has been pressed "a bit further"). - As far as I know, this is normal.
And the "disappointed" expectation can actually be perceived as a braking effect in a subjective sense.

Quote:

However, the effect remained that at full throttle from a low RPM, the acceleration was extremely lean. When the accelerator pedal is then slightly released, the acceleration improves. It's also better if you press the pedal slowly.

This could be related to the load-based strategy, which I sometimes observe on my LDA:
At 1500 rpm with partial throttle, the engine computer allows the boost pressure to rise to approximately 1.2 bar. As you continue, the pressure will be limited to a maximum of 1 bar at a certain point.

Although I don't feel any loss of power, if the LMM is already reporting too little air, the boost pressure could trigger the soot sensor, which in turn reduces the injection quantity . . .

Hello Ulf,

It might be the case that I'm imagining things now, but the fact is that this strange effect (with braking or even just deceleration) has improved since I disconnected the AGR and cleaned the valve. Also, I haven't been driving this car for just yesterday. And the effect of the slight acceleration when quickly pressing the accelerator is exactly the same as the 'braking effect' with (subjective) 'small pedal movements', which definitely occurred abruptly.

What do you think about my strange LMM values?
I re-examined your fault-finding methodology earlier. I was able to get to the exact point, Block10, and the LMM signal. When the engine is stopped, the reading is 1275, and from idle to maximum speed, it's 400-850. There seems to be something wrong, doesn't it?

Greetings
Thomas

PS: There's something about 'target values in the error database'. I think I've seen something like that before. But now it's gone!?! Would I have just printed it out, or could someone send it to me?

Hi,

Regarding the target values: Please refer to the subsequent pages, as all the information is still available there. Or simply search for "Target Value" in connection with your account number.

Best regards,

Jan

meyert wrote:

What do you think about my unusual LMM values?

Hi Thomas

Elevated values above the target are perfectly normal when the AGR is inactive/broken.

Quote:

I just went through your fault diagnostics system again. I was able to get to the exact point, Block 10, and the LMM signal. At that point, with the engine stopped, I got 1275, and from idle to maximum speed, I got 400-850. Something's not right, is it?

not necessarily: the air-fuel ratio (per piston stroke, or better per cylinder fill) is approximately calculated by dividing the LMM signal by the engine speed.
When the engine is stopped, the RPM is zero. And whatever is divided by zero gives 1275

Quote:

There's something about "target values in the error database".

. . . are now available in one of the folders on Rainer's website.

Hello everyone,

clear, with the AGR: Before cleaning, the AGR values were high because it wasn't closing properly. Now they are high because the AGR is out of operation. Therefore, I would have to put it back into operation and measure it again. Then I should see values in the normal range. Perhaps I could draw conclusions about the condition of the O2 sensor. It is already over 200,000 km old and could be responsible for the loss of power. If the entire loss of performance is not just a neurotic figment of the imagination :-).

I have already seen the target values in the folder list. However, compared to the fields mentioned in the troubleshooting guide, the coverage is not very high. I thought there might be more.

meyert wrote:

It's clear that with the AGR: Before cleaning, the values were high because it wasn't closing properly. Now they are too high because the AGR is out of service.

Hi Thomas

If the AGR (AdBlue system) remains open even though it should be closed, the air values will naturally be too low

Quote:

So, ich müsste sie jetzt wieder in Betrieb nehmen und neu messen. Then I should see values back in the normal range again.

Theoretically, yes

Quote:

Perhaps I could draw conclusions about the condition of the LMM. It's already over 200,000 km and could potentially be responsible for the performance loss.

"Weak LMMs typically deliver high throughput = full load, but with low values."
As long as the AGR (at partial load) is active, this can still be compensated for by the control system through a correspondingly lower exhaust gas content. Therefore, "target air values" with active AGR do not necessarily mean "LMM = ok".

Hello Ulf,

Now he's driving again like he used to. I have identified the following problem:

In a previous DIY project, the nipple on the AGR valve regulator, onto which the hose to the AGR valve is attached, had broken off. I had repaired it with a 2-component adhesive at the time. It initially looked quite good. However, it seems that the adhesive is not as temperature-stable as promised, because when I looked at it now, the nipple was broken again and the regulator valve was open at that point (uncontrolled release of vacuum???). In addition, traces of the adhesive can be seen, so it is possible that the effective cross-section of the connection had already been reduced by the adhesive before. Maybe even glue has leaked into the control valve.
In any case, I have now closed the inlet with a vacuum directly in front of the control valve. Now the mileage is back to normal.

Whether the LMM is now working or not, I still don't know. I came up with the idea to check it because I recently drove a Golf 4 with a 130hp TDI engine, which was much faster than my A4 AFN. But the best thing will probably be to forget this experience and enjoy the now-good performance of the A4 again. It's all just a matter of scale:-).

Greetings
Thomas