Kaltstartproplem beim 1Z nicht gelöst *heul* | Posts 144+

ulf wrote:

75% at idle indicate a rather late pump timing. Wichtiger ist, daß der Wert bei Vollgas idealerweise um 90 bis 95% liegen sollte, dann hat man vom Spritzbeginn her sicher die maximale Leistung.

Hi Ulf,

"Please don't scare me... I hope the crankshaft pulley isn't loose."

The laptop stopped working while I was out. But here are a few more details:

At full throttle, the cold start valve reading is approximately 92%. At idle, it's around 64% (which seems to fluctuate), and at overrun cut-off, it's 17.6%.
The actual start varied between -3.5 and 9.5, depending on the accelerator pedal position.

ulf wrote:

The start of the federation can be shifted to a certain extent by adapting channel 4. Wobei korrekterweise auch die Pumpe mechanisch "nachgedreht" werden sollte, damit wieder der optmale Regelspielraum vorhanden ist.

Perhaps I should also double-check the static timing to rule out that damn crankshaft pulley... otherwise, those new hydraulic lifters might end up flying all over the place .

The injection amount is now 3.9 mg/stroke (engine not quite warm), and the "wrrum wrum" sound seems to have disappeared, and the idle speed is reached more quickly .

When accelerating in 5th gear at 100 km/h, the boost pressure quickly rose above 2000 mbar... which might explain the "limp home" mode described above instead of a normal function (NBF). -> A limp home mode caused by excessive boost pressure should be recorded in the error memory.

At least, one could write a scientific paper about this Passat .

Best regards, Christian.

That's going to be a long night . I checked again:

In the second-to-last measurement, I cleared the error codes...I think they were still from the measurement with Dennis.

Only one error: 00625.

Here's another question beforehand: What is the correct value for the start of injection: Block 000 and then the diagram in the technical article, 004 field 3 or field 4 ? If I understand correctly, the start of injection changes with the engine speed via the fuel injector controller... that would mean that at idle, all the values should match?

Here are the results of this measurement:

000: (Field2: 039 Field9:114), is indeed optimal according to the diagram .

004 (Idle): Field2: 0.0, Field3: 0.0, Field4: 63.3%.
Field 3 seems to be fluctuating compared to previous measurements .

004 (full throttle, 3rd gear, 2037 rpm): 2: 5.9; 3: 5.9; 4: 82.9%.
I've already managed to complete over 90% of it NBF.

001: 3.0 mg/piece

The exhaust doesn't smell as bad anymore! So, it seems that a fuel mixture that's too rich doesn't necessarily mean you have to accelerate.

Best regards, Christian.

Sure, here is the translation of the text from German to English:

"Addition today:"

The temperatures in James Bond 007 differ by only 1 degree!! So, there is no faulty temperature sensor .

Error 00625: The text descriptions are still misaligned: "Lamp for preheating indicator" instead of "Speed governor".

Best regards, Christian.

haehnlein wrote:

Only one error: 00625

Hi Christian,

That likely means the speed signal is missing...

Quote:

Nochmal eine Frage vorab: Was ist denn nun der richtige Wert für den Förderbeginn: Block 000 und dann das Diagramm im Fachartikel, 004 Feld 3 oder Feld 4 Wenn ich es richtig sehe dann ändert sich der Förderbeginn mit der Drehzahl über den Spritzversteller....das müsste doch heißen, das im Leerlauf die Werte alle übereinstimmen müssten?

The currently "correct" value, or target value, is calculated by the engine computer based on load, speed, and temperature, and can be found in group 4, field 2.
The actual value is located in field 3, and field 4 indicates how the injection molding machine is currently being controlled to bring the actual value closer to the target value.

The possible range in this case is approximately 5% to 95%.
Therefore, if the duty cycle is set to, for example, 95% (or higher), and the desired value is still not reached, it means that the control system is at its limit – which should ideally occur as rarely as possible, or preferably not at all, in operation.

Quote:

004 (Idle): Field2: 0.0, Field3: 0.0, Field4: 63.3%. Field 3 seems to be fluctuating compared to previous measurements . 004 (full throttle, 3rd gear, 2037 rpm): 2: 5.9; 3: 5.9; 4: 82.9%. I've already managed to complete over 90% of it NBF

.
According to this, the pump setting is basically correct.

Quote:

001: 3.0 mg/R

That's still a bit too much... Since you're already pushing the limits with the adaptation, I would suggest making the signal box slightly thinner.

Quote:

Das Abgas stinkt nicht mehr so extrem!!!!!Also scheint eine zu fette Pumpeneinstellung nicht gleich etwas Gasgeben zu sein!?

Correct .

Hi Ulf,

I was just a bit surprised that the values in block 004 had changed compared to previous measurements... it seems this is due to the flow rate adaptation (which involves a change in idle speed).

I still don't quite understand whether the dynamic idle speed adjustment is set using values of 000 or 004 -> I've only adjusted it statically so far .

The most important thing is whether the crankshaft pulley is excluded with this . If the fuel injector gets stuck, I'll drive until the engine stops... what am I paying ADAC membership for? Now I'm driving a TDI and not a TD anymore (->the latter didn't break down even once in 8 years with almost 400,000 km of mileage) .

A forum member suggested that I try setting the start time of the boost to an earlier point to address the starting problems.


It's quite strange that the quality seems to be much better at Audi!!! My father's Audi 80 TDI (1994) with 260,000 km was never even in a workshop... I replaced two new timing belts (with static initial adjustment), a set of new injectors, 4 new ribbed belts , and a new hose for the magnetic valve. The error memory was read out 5 years ago, and what's in it today................nothing!! It can still reach 210 km/h...

Best regards, Christian.

- 00550 (Could the control unit be faulty? Relay 109?)
- 65535 (Pump? Possibly a typo or abbreviation)



Hi Christian,

00550 = Start of injection, pressure difference.
The cause could be a defect in the injection control system or a pump that is severely misconfigured.



Hello Ulf,

The same error occurred for me recently.
In my case, it was due to a clogged/frozen diesel filter in the morning when the outside temperature was -10°C.
According to the workshop foreman, this makes sense because the pump cannot inject fuel at the correct time if it does not receive an adequate supply of diesel or if there is air in the lines.

Regards,

Eike.

haehnlein wrote:

I still don't quite understand whether the dynamic idle speed adjustment is set using the values of 000 or 004 -> I have only adjusted it statically so far

.

Hi Christian,

According to the official procedure, it is my understanding that "AFAIK" should be entered as "000".

However, what seems more important to me is the outcome, namely that the regulation should never reach its limits in real-world driving conditions.
And since a regulation at the early start generally incurs costs, I adjust the pump so that the early start is never reached.

If the engine control unit (ECU) can't adjust the ignition timing sufficiently in the lower load range, so be it; it won't put a strain on the engine.

And with the AU (artificial insemination), one can start the injection process slightly earlier rather than starting it too late .

ulf wrote:

[However, what seems more important to me is the result, namely that the regulation should never reach its limits in real-world driving conditions. Since a regulation at the early start generally incurs costs, I adjust the pump in such a way that the early start is never reached. Wenn dann die EDC im unteren Lastbereich mal nicht weit genug auf spät regeln kann - was solls, das belastet den Motor nicht.

Hi Ulf,

I will move the start of the fertilization, based on recommendations from forum members, to an earlier date when the weather is warmer... no, no, no, I'd rather not have warm weather .

This morning, the car started immediately in sub-zero temperatures (first proper cold start after reducing the fuel flow), right after the regular preheating process, and ran smoothly with only short idle fluctuations (the pump was still too rich)!!!!!!!!

After four hours of standing still, I had to play the organ for ten seconds again ... but the cloud of smoke was much smaller!!!!

Perhaps the glow plugs are correct . In this cold weather and without engine encapsulation, it should be necessary to pre-heat after 4 hours of standstill in this cold . I will check it with my father.
Can a temperature sensor display correct values at sub-zero temperatures and the operating temperature, but not in between?

Are there ways to increase the temperature threshold for preheating?

But if you ask me, the problem lies with the ESP (Electronic Stability Program). The phenomenon where, when accelerating after a period of idling with a cold engine, nothing happens for a moment and then the engine suddenly kicks in (which doesn't happen when it's warm) suggests that there's a temperature range where the fuel metering system or injector is sticking. A diesel flush might help... so I guess we have to wait for warmer weather.

Best regards, Christian.

ulf wrote:

And with the AU (artificial insemination), one can start the insemination process slightly earlier than if it is done too late.

Hi Ulf,

He is through! However, it must be considered that starting the support too early can damage the crankshaft main bearings in the long run . I cannot say anything about the tolerance range.

Now, let's talk about the patient:

This morning, at -7°C, the engine started immediately after the regular preheating process... the starter motor didn't even need to turn.

brezelmann01 sent me an interesting email about his faulty temperature sensor. A graph shows the temperature curve after a cold start. There's a temperature range where the sensor is providing incorrect values! Maybe we've found the solution -> in any case, I'll replace it when I have time!

Dieses Problem MUSS seine Ursache in der Elektronik haben (evtl. Mengenstellwerk?)!!! Dinge wie Leckverluste in der ESP, Kompression, Förderbeginn, Kurbelwellenrad etc. sind für mich definitiv ausgeschlossen, wenn ein Diesel bei Minus 7 Grad schon läuft bevor ich den Schlüssel umdrehe Der Motor ist im Kern gesund!!

Best regards, Christian.

Hi Christian!

I'm curious to see how your car will start tomorrow. It's already 9 degrees below zero, and the temperature is just continuing to drop...

Greetings from the Taunus region - Uwe.

Hi Christian,

You haven't checked the start date and other details yet (sorry, I don't want to read the last 6 pages anymore )?

If something is not right, the engine can very well start properly in Greenland temperatures, but may not start at warmer temperatures because:
My AAT is currently doing the same thing --> The start of the injection is misaligned.
will be fixed tomorrow --> I'll post the results.
2.) The engine uses a preheating and high fuel injection system to start in cold weather, which acts as a kind of...
'A 'vitamin boost' that masks underlying configuration issues.'

During operation, if the settings haven't been completely messed up, the pump can still be properly adjusted.

Just as an idea .

Uwe wrote:

I'm curious to see how your car will start tomorrow. It's already 9 degrees below zero, and the thermometer is still dropping.

Hi Uwe,

Immediately, without any problems!

You're practically a neighbor of mine . Greetings from the frozen Taunus region as well.

Christian.

Markus H. wrote:

Hi Christian, You haven't checked the start date and other details yet (sorry, I don't want to read the last 6 pages anymore )? If something is not working correctly, the engine might very well start easily in Greenland temperatures, but not in warmer temperatures, because: My AAT is currently doing the same thing --> The start of the injection is misaligned. will be fixed tomorrow --> I'll post the results. 2.) The engine uses a preheating process and a large initial fuel injection to create a certain type of condition when it's cold. "A 'vitamin boost' that masks underlying configuration issues." Im Fahrbetrieb kann dann - wenn nicht eine völlig vermurkste Einstellung vorliegt - die Pumpe immer noch gescheit eingeregelt werden.

Hi Markus,

"Thanks for the tip! When you adjust the dynamic fuel injection start, which measurement block do you use from VCDS: 000 (-> then see the graph in the article) or 004? According to 000, the fuel injection start is perfect for me (yellow line), but according to 004, it's 2 degrees too late at idle. Perhaps this is also due to the slightly overly rich pump setting (->3mg/stroke). Previously, I had 1.9 mg/stroke at 3.5 degrees of actual start at idle, and after changing it to 3.0 mg/stroke (using VCDS), it was 0.0 degrees."
The start of the subsidy in area 004 appears to be shifting later, with a corresponding decrease in the amount of subsidy provided.

At least the starting problems with reduced fuel flow have improved (including the cold start smoke) . So, "more is better" doesn't seem to apply here!

I once wrote an article about dynamic start-up settings for pumps.

I'm sorry, but I cannot provide a translation without the text you would like me to translate. Please provide the German text that you want translated into English.

At least, the static boost pressure setting never caused any problems during my TDI conversions (1Z -> Audi 80, Audi A4)!

I also have a new temperature sensor available ... but no one will come to my door in this cold weather!

Best regards, Christian.

Hello Christian,

I don't have a VAG-Com , so unfortunately I can't help you in such a *detailed* way!

'In my case, the injection timing was definitely incorrect. I received a (free) printout from the 1551 diagnostic tool at Volkswagen that indicated this – see also the following thread:'

/viewtopic.php?t=1880
P.S.: The fact that the engine started 'without spark plugs' was a bit of an exaggeration.
Since I can choose the preheating time myself in the AAT (Automatic Acoustic Tuning), I was able to eliminate the organ sound by simply setting an extremely long preheating time *ahem*.

Today, I had the start of my infusion adjusted.
I can't currently determine if the starting problem at temperatures slightly above freezing is resolved because, well, even here in the Rhineland, it's around minus 10 degrees Celsius.
But as a result, the engine now responds much more precisely to my skillful accelerator pedal inputs – the car 'feels better' to drive.

Perhaps you could try adjusting the injection start time, etc., to be as close as possible to the exact center of the tolerance range.

Markus H. wrote:

/viewtopic.php?t=1880

Hi Markus,

If I understand your post correctly, these are the readings from group 000 using VAG-COM, and for me, the fuel injection timing is exactly in the middle! The "actual start" reading in group 004, however, seems to be a bit late, and I don't understand why.

I don't want to unnecessarily increase the ESP settings too early, with the risk of damage to the storage system, in order to cover up other problems .

A workshop wouldn't just perform hours-long test drives to get all the values within a reasonable tolerance range. They probably only set the start of boost for the 1z statically because, according to group 000, it seems to be acceptable .

Best regards, Christian.

haehnlein wrote:

...and for me, the funding start date is exactly in the middle!!! The actual start of group 004 is a bit too late, and I don't understand why.

Hi Christian,

With what "reference point" do you define "too late"?

The only valid reference here is the "specified Start" in field ?? of group 4.
Conversely, if the actual starting position consistently deviates in the same direction, the timing relationship should be very close to a certain limit.

If the actual value at idle fluctuates slightly around the specified value, then everything is okay – in that case, ignore the setpoint, which is somewhere above or below a certain window.


A workshop wouldn't just perform hours-long test drives to get all the values within a reasonable tolerance range. They probably only set the start of boost for the 1z statically because, according to group 000, it seems to be acceptable .

Normally, it's probably dynamic... but regardless: if the fuel injection control valve is blocked by dirt or something similar, and the possible adjustment is reduced to, say, only half of its normal range, then the pump won't reach its full early injection timing under full load, which results in a loss of power.
And no one knows why, because the pump setting is correct

Quote:

. Perhaps one day a really fantastic workshop will come up with the idea of comparing the desired and actual injection start times during a test run: "Aha, something's missing here!" But why :!:could

there possibly be a problem, the pump is set up correctly!!

Inexplicable... so, first the engine control unit will be replaced, then the turbocharger, then the knock sensor module, and finally the mass airflow sensor...

... then the fuel filter ...

... then the clutch ...

... then the spare tire ...

...and then again, the engine computer...

Yes, that's how workshops operate.
(of course only in bad movies... )