TDI AFN Engine: Starting Issues, White Smoke, Misfire

Dear diesel mechanic enthusiasts,

I've been having some serious problems with my Golf 3 over the past two days, and I wanted to share my experience and put some questions up for discussion.

It's a 20-year-old Golf with a TDI engine (AFN). The mileage is approximately 460,000 km. It still has the original turbocharger (I'm quite proud of that). No chip tuning, everything is original.

On Monday, I couldn't start the car. I cranked the engine repeatedly, and a huge cloud of white smoke came out from the back. Eventually, the engine started, and the car seemed to run normally. Because I was in a hurry, I didn't think much of it at the time and drove off. In my opinion, the problem was definitely with the glow plugs, because the car otherwise ran perfectly normally, without any problems or loss of power.
In the afternoon, I checked and discovered that the power strip supplying electricity to the glow plugs was completely rotten and crumbled into small plastic pieces just by looking at it. This confirmed my suspicion – that was the problem. I needed to get a new power strip and remove the glow plugs to inspect them. I planned to do that over the weekend.
Yesterday, of course, I had the same problems again with smoke during startup, but then I was able to drive normally for about 15 km. At some point, I noticed that the car briefly jerked when accelerating out of the idle range. I could always see a white cloud of smoke in the rearview mirror. Otherwise, no other problems.
When I finally arrived at work (about 30 km later), the car started to sputter and smoke even while idling. It seemed like the engine was about to fall apart. I immediately turned off the engine, and in the afternoon, I had it towed home.

During my lunch break, I searched for information in this forum and also found some ideas for troubleshooting.
At home, I first ran a diagnostic check using VCDS. This revealed that the EGR valve (N18) was defective. I removed it and tested it, and sure enough, it had a short circuit. It was always activated, which meant it was always keeping the EGR valve open. I then wondered if the defective valve could really have such serious consequences, so I disconnected it and blocked the hose leading to the EGR valve with a screw. Then I tried to start the engine: nothing. It wouldn't start, not even a brief misfire or anything.

Okay, so I thought I'd check the timing. And guess what! It was actually misadjusted. It seemed to be advanced. So, I worked on it until 11:00 PM yesterday and corrected the timing, hoping that the engine hadn't suffered any further damage. Then I put everything back together and started it: The engine started right up.
This morning, I drove to work in the car (with the EGR valve disconnected), and it seemed to be running normally. I noticed that it might be a little louder than before, almost like a knocking sound.

I have already ordered the N18 valve. Otherwise, I will observe whether the engine has sustained any further damage.

Moral: When experiencing starting problems, always investigate the root cause thoroughly and don't simply blame the glow plug system.

Questions: Can a faulty EGR valve really cause a change in valve timing? Or did I have two separate issues at the same time?
Should I disassemble the engine to check for further damage, or is there another way to determine if it has sustained any additional damage?
How long can you drive with a disconnected EGR valve (with the valve always closed) without damaging the engine? I understand that the check engine light will be on.

Thank you for reading and for your feedback.
Greetings from Andalusia.

Kiko.

Hi,

My thoughts:
-Glow plugs have nothing to do with starting problems in TDI engines. At least, not until temperatures drop to zero degrees Celsius: I was able to start my 1Z with a faulty glow plug fuse without any problems. It ran a bit rough, but it ran.
- Timing: What does the KW screw look like?

Best regards,
Adrian.

Hello Adrian,

Thank you for your reply.

I should have known that the glow plug system wasn't the problem. This topic has been discussed many times in this forum. I still have an old Mercedes /8 diesel, and it requires a preheating time of at least 1 minute. Perhaps that caused some confusion.
Since I didn't notice any other problems initially, the matter was settled for me.
Therefore, to everyone: If you experience any difficulty starting the engine, always determine the cause completely and do not simply drive off.

Regarding the crankshaft pulley bolt: I looked into this issue when I bought the car (year 2007). To my knowledge, the new bolt with the higher strength is installed. I have never replaced it myself. I have always done the timing belt replacement myself, but I have never replaced the crankshaft pulley bolt, even though a new bolt is always included in the repair kit. I always assumed that it is only necessary if the older bolt is installed.
Yesterday, I couldn't detect any loosening, and the timing belt pulley on the crankshaft looked normal.

Regards,

Kiko.

Hi,

Quote:

Questions: Can a defective EGR valve really cause a change in valve timing?

No. Please make absolutely sure that everything is securely fastened on the timing belt system, including the tensioner pulley, timing belt gears, etc.

Best regards, Rainer.

Hello Rainer,

Thank you for your comment.

Now that my question is standing alone, I can't imagine how the valve timing could change. I had the thought that a constantly open EGR valve might build up pressures that work against the actual piston movement, thus counteracting the rotational force. However, since gases are compressible, this is very unlikely.
I've experienced similar severe vibrations before due to a faulty alternator one-way clutch. In that case, the tensioner pulley for the serpentine belt failed, and the engine shook and rattled like crazy, but the valve timing didn't get thrown off.

I'm going to open the front panel again and carefully check everything.
By the way, the timing belt was replaced about a month and 5000 km ago. I didn't replace the stud for the tensioner pulley (there's always a new one included in the kit). It might be causing problems. It looks like a stretch bolt. Perhaps it's reached the end of its "stretch" capacity.

I'll get back to you.
Thank you again for your time.

Regards,

Kiko.

Brief update:

Since I didn't feel like checking the injection start yesterday evening after assembling it (it's still evening here in Andalusia at 11:00 PM), I made up for it today. And there it was: I had something around 15 degrees before top dead center. There was even an error message displayed.

Adjusting the pump by rotating it didn't yield the correct value. I was getting 8 degrees before top dead center. So, I did something that might not be entirely professional: I rotated the pump back to its original position, then loosened the timing belt and very carefully rotated the ESP timing pulley by one tooth.
Everything was reassembled and checked with a timing checker: the engine ran perfectly, and the fuel injection timing was within the specified limits. The knocking sound was gone.
Then I took it for a short test drive around the block. At first, the car ran perfectly, but after about 1 kilometer, the problems started again: First, it jerked when accelerating with white smoke, then it sputtered, and I barely made it home. Now the engine won't start again, and it's smoking like crazy.

Today, I don't feel like it anymore.

I will completely disassemble the timing belt again tomorrow and check everything carefully once more. Somewhere, there must be a component that is altering the timing. Fortunately, it's only affecting it to the extent that a major engine failure hasn't occurred yet.
I will report...

Good evening.

Kiko.

Sounds like a potential keyword stuffing issue! Please review this carefully.

Hello... this is about the crankshaft pulley bolt. I think there's something loose in your timing belt system, which is why you have to "readjust" the belt again. If you still want to keep the car, now is the time to stop and not start the engine until a thorough inspection is done.

"Yes, that's definitely the KW screw."

Even the "new" screw can come loose, which also happened to me with my AFN.
It's possible that the faulty Lima freewheel is contributing to the problem, but that junk is already old.
It was quite loose. I drove with the fixed camber plate for about a year before the KW screw came loose.
was.
You can also permanently disable the AGR (automatic gearbox) while it's in the closed position.
Not even a single error entry in the AFN logs.

Greetings,
Dieter.

Hello everyone:

Thank you for your comments!

Just to clarify: The faulty alternator clutch has been defective for about a year. This isn't directly related to the current problem, but it might be a delayed consequence.

I'm now quite certain that the problem must be either with the screw or the timing belt pulley. I've always assumed that a loose screw or a defective timing belt pulley would be immediately noticeable. I turn the engine precisely at that screw, and I can't see that it's loose in any way.
Perhaps the screw is actually quite tight, but the groove or spline on the crankshaft is worn out.


It's all useless! It looks like the screw either has to come completely off or be tightened all the way.
Then we'll know more.

Do I need a special crankshaft holder, or is it sufficient to put the engine in a high gear, which will block the crankshaft?

Best regards from the Costa del Sol (it's supposed to be beautifully warm today, which makes working on projects even more enjoyable...).

Kiko.

You will likely need a new crankshaft pulley in 99% of cases, and a milling cutter for the crankshaft stub, because it is certainly no longer flat, so a new pulley will only provide a temporary solution.


http://www.dieselkontor.de/product_info.php?products_id=3500&language=de

Dear Rüdi:

Thank you for your feedback as well.

Yesterday, despite the heat, I worked on it again. I just didn't want to bother everyone in this forum again.
Since I'm already online, I might as well give you a quick update:

I was also absolutely convinced until yesterday that there was something wrong with the timing belt pulley or the "killer screw." So, I took everything apart again (I think I could now replace the timing belt in record time).
"So, I took a very close look at the area below. I couldn't find any looseness or problems with the screw, no matter how hard I tried. I blocked the motor using the transmission, and my wife pressed the brake pedal from inside the car. Then, I pretended to loosen the KW screw, but nothing was loose. I also checked to see if the rotations I was making on the front were the same on the rear, specifically on the flywheel ring gear. When I turn the front, everything rotates correctly on the back. There's no play. That's why, at the last minute, I decided not to open the screw after all. It's a classic case of 'never touch a running system'."
I then assembled the parts that were absolutely necessary for the engine to run, leaving out things like the belt guard and the other cover for the auxiliary components.
Then the engine was adjusted to the correct timing and allowed to run for a short time. The timing belt pulley is also not warped. The engine runs smoothly - at least for those few minutes.

Ich habe dann noch die Spannrolle geprüft und mit der alten verglichen. Die hatte ich noch in meiner Schrottkiste liegen. Alles identisch. Den Zahnriemen habe ich mir angeschaut. Alles ok. Die Nebenwelle für den Antrieb der Unterdruckpumpe habe ich auch unter die Lupe genommen. SEverything is fine.

After that, I reassembled everything completely, and I discovered a potential error:

When assembling the timing belt, I always tightened the large bolt on the camshaft to 60 Nm first, and then the tensioner. I then manually rotated the engine completely five times (because I wanted to see if anything had shifted), and I noticed that the valve timing had shifted slightly.
After I opened everything up, adjusted it, and then tightened the tensioner pulley and then the large screw on the camshaft, I couldn't notice any changes. Even after completely turning it over 10 times (if you're working on something, you don't need to go to the gym).

So, there I was: On my right side, a little devil was whispering in my ear, saying: "Dear Kiko, I assure you that the crankshaft pulley bolt and/or the timing belt pulley are defective. Take everything apart!"

On my left side, a little angel whispered in my ear and said: "Dear Kiko, don't worry anymore. You probably made a mistake during the last ZR belt replacement, and after about 5000 km, something probably got misaligned."

So, I put everything back together and started the engine (and it vibrated). The engine ran perfectly. I could hear knocking. I ran a timing check, and the injection timing was at 3° to 4° BTDC, which is a bit early.
Ich bin ein paar Mal um den Block gefahren, habe absichtlich etwas grobere Lastwechsel verursacht. Everything seems to be okay.

Today, I still drove to work using the emergency starting method. I'm still not completely confident about the situation. This weekend, I'll properly adjust the fuel injection timing and then drive around town, staying close to home.

What do you think? Could it be that the defect in the crankshaft pulley is so small that everything seems perfectly secure, but there's still a tiny amount of play? Perhaps it only becomes noticeable when the engine is very hot.

Wishing you a lovely weekend.

Kiko.

Hello, there seems to be a problem... The KW wheel is twisting on the shaft. This is also the reason why the shaft stub is getting damaged. If you were to twist it even slightly with a wrench, the motor would be ruined.

Dear Matthias,

Thank you for the information.
I wasn't aware of that.

I will remove the timing belt pulley.

Best regards,

Kiko.

The gear can only rotate on the shaft if either the screw is loose or the fit is worn out (or both). In the latter case, the screw must have been loose at some point previously.
Replace the screw to give your stomach some relief, and maybe also the spur gear, but for now, avoid using any planers or similar tools.
You can upload images of the inside of the valve guide and the camshaft here.
Use an original screw and install it according to the regulations. You can find the data for the AFN in the RLF documentation.
I would start by properly adjusting the fuel injection pump mechanically, and then check the target and actual values for the electronic fuel injection control (EZP). Keywords: EZP control, valve for injection timing, internal pressure, possibly also a needle lift sensor.
hg
Herbert.

PS: As a workaround, you can attach a piece of steel strip or something similar to the front gear using two screws. However, the loosening and tightening torque is not good for the KW (likely referring to a specific component), the clutch, and the transmission.

@Herbert: If the wave pattern deviates from the original even by the slightest amount, the entire thing will only last for a few thousand kilometers, and we'll have to start all over again.

The screw doesn't necessarily have to be loose, and the fit you mentioned only serves as a guide; it doesn't secure the wheel hub to the driveshaft. The wheel hub is secured solely by the surface pressure on the intact contact surfaces, which is created by the correctly tightened driveshaft screw.

Previously, the problems were caused by a screw that was too soft being used from the factory (8. was used). Today, the problem is due to the age of the screw and the fact that it slowly fatigues, causing the KW wheel to start moving on the shaft. If it has even moved minimally, then that's it.