Adjustable Injection Start: Not Possible

Hi.
We have a problem with our Audi A4 B5 AFN from 1996: the dynamic injection timing cannot be adjusted.

The following actions were taken.
The ABS unit was leaking. I removed it and took it to a Bosch service center in Estenfeld for repair, sealing, and inspection. I thought that after 270,000 kilometers, the old girl deserved it.
"I got the pump back yesterday. 'Everything is fine. The pump has been sealed and adjusted.'"
Today we installed the pump. The timing is perfect. The alignment and fitting of the pump were excellent.
Key turned, engine started immediately.
In idle mode, the value in field 2 of module MWB 000 is approximately 53.
In the default setting, I can almost fully adjust the range, but I don't get any values above 000. It always displays that the injection timing is too early to show a value.
The engine runs smoothly when I set the fuel injection timing to its "normal" setting.

What I've noticed.
The fuel injection start timing (MWB 004) is within the expected range and functioning correctly.
But:
The start of injection, even at idle, is between 1.3 and 1.8 degrees before top dead center (BTDC).
If I increase the speed until it reaches 2100 RPM, the value goes to -2.6° relative to top dead center (TDC).
Above 2100 RPM, the injection timing shifts back towards before top dead center (BTDC).
Bei 4000U/min bin ich schon wieder bei 10°v. Off-topic.

Could it be possible that the pump is not working properly?
Or maybe I made a mistake.

The valve timing is definitely correct.

Do you have any other suggestions?
There are many threads on this topic, but I haven't been able to find a solution using them.

I also have a log file from MWB 000,001,004, starting from idle up to approximately 4300 RPM.

Regards,
Bolzer.

Okay, I'm ready. Please provide the German text you want me to translate into English. Just paste it here, and I will give you the translation.

Are you sure that you have also set the "TimingChecker" to AFN mode?

Hi there,
The measured values are independent of the model, but the tolerance range depends on the engine.
It appears that the ESP is significantly out of phase, which explains the "000" reading in MWB2. This also aligns with the fact that I'm seeing a duty cycle of 78% in the LL (Low Level) signal, whereas I only see 3%.
The EDC system manages the situation reasonably well under the shown conditions; therefore, outside of the basic settings, the "normal" values are displayed, and the adjustment is made early on based on engine speed.
hg
Herbert.

Hi.
The timing checker was set to the AFN setting.

OK. Du meinst dass der Spritzbeginn deutlich zu spät ist. Ich werde am Montag nochmal eine Grundeinstellung vom Spritzbeginn machen. Let's see.

Regards,
Bolzer.

Hi there,
What static value did you set when installing the pump?
hg
Herbert.

Hello,

Herbert wrote:

It seems that the ESP is significantly out of phase, which explains the 000 value in MWB2. This also aligns with the fact that I'm seeing a duty cycle of 78% in the LL, whereas I only see 3%. The EDC system manages the situation reasonably well under the shown conditions; therefore, outside of the basic settings, the "normal" values are displayed, and the adjustment is made early on based on engine speed.

Conversely: The higher the water level (TV), the lower the internal pressure of the pump, which leads to a later start of the system (SB).
To me, it looks more like the pump is slightly too advanced in its timing. In low load conditions, it's not able to retard the injection enough, even though the duty cycle is at its maximum (94%).

(SHOULD) (IS) TV
Certainly! Here's the translation:

"∞v.OT ∞v.OT %"
1. 3 1.8 78.5
1. 3 1. 8 94


Best regards, Uli.

Negative "low" duty cycle percentage means high valve activation. In the basic setting mode, 0% corresponds to full activation.
Try it out, for example, by starting the actuator diagnosis - proceed to test the injection start valve, then exit the diagnosis and view parameter group 4 (PG4). Afterwards, you must of course complete the actuator diagnosis.
Or search the forum. There was an article there about how to adjust the ESP for optimal control values. At high power output and thus earlier injection timing, it's best not to exceed 95% so that there is still some reserve.
hg
Herbert.

Hi.
I'm sorry I didn't reply today.
Family outing. It's nice to not have to think about cars for once.

Ich hatte den Förderbeginn mit der Messuhr eingestellt. Value approximately 0.7.

However, I had to return the watch to the service center again. Hopefully, I'll get it back sometime.

However, what makes me a bit uneasy is the fact that the pump was disassembled and cleaned by a professional company. I'm not sure what they did to the pump during that process.

Could the problem possibly be related to the pump? I wouldn't want to spend a week troubleshooting something that might not even exist.

But please don't misunderstand me. I'm not trying to imply that the workshop made a mistake, but rather to rule it out as a possibility.

Regards,
Bolzer.

[Translating...]

Moin,
die 0,7 mm Hub bei OT kommen schon hin, vorausgesetzt bei der Messung sind keine Fehler gemacht worden (ZR Spannung, Vorspannung der Uhr, Nulleinstellung, OT durch Drehen der KW in Laufrichtung anfahren, Ablesen des OT). Ich hätte sonst vermutet, daß das Riemenrad der ESP beim Auflegen des ZR um einen Zahn daneben lag.
Prüfe mal den Verstellbereich der Spritzbeginnregelung mit der Stellglieddiagnose, d.h. diese aus dem Testschritt "Ventil für Einspritzbeginn" verlassen und MWB 4 beobachten. Das Ventil wird für ca 30 s zwischen minimal und maximal gesteuert. Welche Werte siehst Du da für Soll, Ist und Tastverhältnis?
Ansonsten die Spritzverstellung mal unter wechselnden Lastbedingungen mit VCDS (Logfahrt) und Spannungsmessung prüfen ( https://community.dieselschrauber.org/en/viewtopic.php?t=3195 ).
hg
Herbert

Hello,

Herbert wrote:

Negative - a "low" duty cycle in % means high valve activation. In the basic setting mode, you have 0% = full activation.

Oh, I totally messed that up ... Sorry, you're absolutely right. High TV adjusts after early, not after late.

Best regards, Uli.

Hi.
He is running.
Ich hab den Adapter für die Messuhr leider diesmal nicht bekommen. But it doesn't matter.
I first marked the entire adjustment range of the pump and then turned the pump to the first third of its range.
Surprisingly, the values matched quite well.
And most importantly, the default setting also had a value of 012.
Slightly better adjusted, let the engine warm up, and then set the timing precisely using the TDI timer.
Now the values match.

I don't know what I did wrong the first time, and I can't figure it out anymore. But the start of the injection was definitely set too late.
I know what my second mistake was. I wrote that I had tested the entire adjustment range. Unfortunately, I only used the adjustment range allowed by the fuel injection lines. I have to admit that I didn't loosen the lines, which meant the adjustment range was too small, and therefore I didn't reach far enough into the early injection phase.
A small, stupid beginner's mistake. I am also ashamed of it.
It's very frustrating because I'm already replacing the timing belt for the third time on this engine, and I've also replaced several of these pumps before.

Here are the logs for the current configuration.
One is set to the default setting, and the other goes from idle up to 4000 RPM.
The duty cycle is still around 55% in the LL (likely referring to a specific component or system).

Thank you very much for your help.

Regards,
Bolzer.

Hi there,
What surprises me:
- Under normal operating conditions, the injection timing at idle is regulated to be significantly earlier than in the basic setting mode, approximately 1° before top dead center (TDC). With a warm engine, it should be later.
- With a slightly elevated lambda value, the injection start timing suddenly shifts to top dead center (TDC).
At first, I thought it might be related to the pressure ratio in the LL (likely referring to a low-pressure system), possibly due to something with the pump's internal pressure, but the actual values are following the setpoint values.
hg
Herbert.