Audi A4: Won't Start - Troubleshooting Guide

Hello dear community, I hope you can help me.

"I've owned a 1.9L AVF engine for the past 4 weeks. During the test drive, the car wouldn't start. The workshop that sold the vehicle on behalf of the customer said that the battery wasn't in the best condition due to it having been unused for some time."
The car started with a little sputtering after being jump-started.
Since I wanted to be absolutely sure, I took the car to the TÜV (German technical inspection agency) for a check-up. The results were good; everything seemed to be in perfect order. I bought the car two weeks ago and have been happy with it so far.
The car suddenly wouldn't start one morning. The starter motor was working fine (the battery had 12.7V). We let the car roll downhill and it started. We turned the car off and tried to start it again, and it started right up on the first try. After two days of no problems, the same thing happened again...
The car is trying to start rolling but without success. A tow truck arrived, it cranked for 30-40 seconds and the engine started very roughly. The tachometer suddenly showed 1400 RPM, but it also runs at its normal 820 RPM. The tachometer doesn't move smoothly, but jumps erratically (suspected air leak).
"The lines and the pre-pump were checked and are working correctly. At my request, the diesel filter and the fuel pressure regulator were replaced because I had already read about problems with the fuel pressure regulator here."
3 days of rest and then the same thing all over again...
A woman called and said her car wouldn't start (it started briefly, then immediately died). Since I was at work, I couldn't do much to help, so I went directly to the hardware store and bought some starting fluid (Startpilot).
Everything was exposed and sprayed with Starter Fluid (5 shots), and then it started right up.
On Monday, I drove to the workshop and made an appointment for Friday to further investigate the problem. They said there were no error codes, which is like finding a needle in a haystack, and it would be better if there was a clear error. On Monday evening, I hoped the car wouldn't start, and unfortunately, that's exactly what happened on Tuesday morning. I called a tow truck to take the car to the workshop without attempting to start it.
Being as impatient as I am, I drove there every day to see how things were progressing. On Thursday, I received a call (from someone who had spoken with my wife) saying it was likely the tandem pump. Since I don't like hearing the word "likely" because it sounds like they're just going to try a quick fix, I naturally went straight there.
He started explaining to me and confirmed that it was likely the pump in 90% of cases.
He showed me that he had blocked the return line leading to the fuel pump with a screw, and that the car starts normally like that. As soon as he reconnects the return line, the car won't start.
"Replaced the tandem pump, picked up the vehicle yesterday, and this morning I saw the same thing, but in green. The car wouldn't start, so instead of waiting another 4 hours for a tow truck, I grabbed a can of starting fluid and, after spraying it in seven times, it started right up."
Drove the car to the workshop; he wants to see if he can make it by Monday and take a look.

Here are a few suggestions from my side, in case they're helpful:

The error memory is empty; there's nothing in it that could be related to the problem.
Replaced the diesel filter and a toad.
Replaced the tandem pump.
When turning the key, the preheating symbol briefly illuminates (relay 109).
The vehicle has full power.
All PD elements are within the specified range.
Harness for PD components, 4 months old.

The log entry "Rule Limit Exceeded" appeared after a long highway drive, and the VTG (Variable Transmission Gear) stopped working properly.
Unfortunately, I don't have a current log because, as mentioned before, the vehicle is currently at the workshop.

Thank you in advance, and I hope you can help me.

Best regards,
Florian

Hi,

Unfortunately, several modifications have been made that are not good for the car.

1. Starting spray used on the diesel engine. Do not use it; see the search function for reasons why.
2. Lubricating the VTG (variable turbine geometry) makes everything worse. When the turbocharger gets hot, the lubricant turns into carbon and causes even more problems. For a VTG that is sticking: The search function provides endless information and also the solution, which is to disassemble the turbocharger or at least move the VTG until it no longer sticks (VCDS basic settings, control module 10 or 11).

The starting problem can be caused by several things, including an unreliable power supply from the engine control unit (check relay 109), air in the fuel system, and insufficient fuel supply (are you already working on this?), a faulty coolant temperature sensor, too low starter speed when hot, and a stuck throttle valve on the intake manifold.

Best regards, Rainer.

Hello Rainer, thank you for your reply.
Yes, I know that using the "StartPilot" is not a good idea.
Regarding the VTG issue, I have already used the search function and read through the relevant information. I will look into it again once I have the vehicle back. (Thank you for the reminder.)

"I would actually like to rule out the relay, because I can access the control unit during the starting attempt, and the glow plug light also comes on briefly. Four mechanics are at a loss, and I'm an amateur when it comes to diesel engines. It's not a Golf Mk2, which is what I usually work on; it's just different technology."

Best regards,
Florian

Have the O-rings on the PD elements already been replaced?

Okay guys, let's not be too hasty. Let's leave the injectors where they are for now. It's a more significant procedure.
"This can be checked. There is a test port on the tandem pump. You can connect a pressure gauge called VAS 5187 there. If the target value is not reached, but it is achieved after pressing the return line, then the sealing rings are likely the problem. In module 023, you may also be able to see which cylinder is affected. The workshop should have done this work right away. That way, they wouldn't have to replace tandem pumps unnecessarily."
Okay, I'm ready. Please provide the German text you want me to translate.

Let the engine cool down and read the values in the MSG MWB7 module. These are the readings from various temperature sensors. Check if they are reasonably plausible and correspond to the ambient temperature, especially the one for the coolant.

Then disconnect the common rail injector connector at the cylinder head. While cranking the engine, monitor the RPM using VCDS. The starting RPM should be around 280 RPM.

If that doesn't help, I would measure the compression. Does it have power, though?

Koala wrote:

The pre-charge pump has been checked. Best regards, Florian

.

This is another one of those candidates. Is the flow rate being measured? Did they just listen to see if the pump was humming or something?

According to his description, the diesel engine is leaking.

He already replaced the tandem pump and the "Knackfrosch" (likely a specific part).

"Since the pressure of the tandem pump is measured at 4000 RPM for most cases, and only a target value within that range is specified, I don't think much can be measured there. Also, the workshop supposedly disconnected the return line, and then the engine started."

The issue with the crankshaft speed usually only manifests when the engine is warm.
Regarding the temperature sensor, it would always need to display values above 70°C for that to work in conjunction with the starter motor speed. Even then, it eventually starts up on its own, without needing the start pilot.


It also has over 200,000 kilometers on it. So, I'd suspect the O-rings.
You could try placing the gas cap on top of the filler neck without screwing it on before parking the car overnight.

chli1976 wrote:

Since the pressure of the tandem pump is measured at 4000 RPM for most units, and only a target value within that range is specified, I believe there isn't much that can be measured beyond that.

For this type, the target pressure is 3.5 bar at 1500 RPM and 7.5 bar at 4000 RPM. Additionally, the internal tightness of the tandem pump can be checked using another device (VAS 1687). Everything can be ruled out when it comes to the tandem pump; it should not be replaced without a proper diagnosis.

chli1976 wrote:

It also has over 200,000 kilometers on it. So, I'd suspect the O-rings.

They don't necessarily have to be leaky just because of the mileage. But okay:

Please do the following when you have the car back:
"Warm up the engine, log in to Msg, and record parameters MWB 13-18-23 while idling for about 30 seconds. You can also briefly accelerate the engine. Then, please upload the resulting CSV file here. We'll take a look."

Hello you two, please don't argue about my problem.
First of all, thank you for your answers and help.
The car will be back at my place tonight, and tomorrow morning I'll read out or record the data blocks (hopefully I can manage to record them, as I haven't done that before).
The workshop wants to replace the catalytic converters based on suspicion, so for this workshop, the case is closed, and I will pick up the car without any further comments.

Now, let's get to the actual problem:

The car started without any problems on Friday after the tandem pump repair. However, when I tried to drive it away on Saturday morning, it wouldn't start again. I had it towed back to the workshop. On Saturday evening, I went to check on it, and it started immediately. "Good," I thought, "I'll go get some bread on Sunday morning and then stop by the workshop to see if I can get lucky." Sure enough, it started right away. This morning, before work, I drove it back to the workshop and tried again, and it started immediately there as well.
The only difference between the workshops and me is that, with me, it's actually always standing with its snout pointing upwards. But during the time when it started easily in the mornings, it was positioned like that. Unfortunately, I don't know if the workshop measured anything. In this regard, I probably have to trust the statement of the workshop foreman, but unfortunately, I wasn't present when he did it.

I really wanted to avoid using O-rings and PD elements as a last resort.
The starting speed at which it fails to start is 280 RPM, which I have checked before.
The values of the PD elements can be found at the very bottom of the log file, as indicated above.

I will complete MWB 13-18-23 tomorrow morning (if I can get the recording to work).

My approach from my side would be as follows:

Remove the diesel filter and the "Knackfrosch" (presumably a specific type of component), replace the filter with one that doesn't have a "Knackfrosch," connect the return line, and additionally install a check valve in the supply line. (Tip from a T3 owner).
Check the hose running from the tandem pump to the brake booster. I read about someone else having the same problem on another forum, and it turned out to be due to that.
Perhaps it would be a good idea to simply replace all the diesel lines; they shouldn't cost too much, and I should be able to do it myself.
I received one tip from a master technician at Audi, but I'm not sure if it's true. Apparently, there might be another tandem valve inside the EGR system that isn't visible.

Apparently, I'm not the only one experiencing these problems, but somehow the solution isn't readily available anywhere. I would like to finish this thread so that it might help other people as well.

Best regards,
Florian

We're not arguing; we're just sharing our experiences.
Steffen probably sees it exactly the same way.

Why do they want to replace the PD elements? I've only had one defective one in the last 10 years (a Bosch), while Siemens seems different.

You can try using a check valve, but I once tried that on an Audi and it didn't work until I replaced the piston rings.
It was a used check valve, but it blocked the flow in one direction.

"Well, as I already mentioned, it's based on suspicion. They talked about the O-rings over the phone, and during that process, they could also replace all the PD elements, which would actually eliminate the problem. For them, it seems like a quick way to make money; they just swap parts one by one, regardless of whether they are faulty or not. Eventually, the problem will go away."

I'll try to record the mileage tomorrow morning and post it here. I'll also try parking the car with the front end facing downwards to see if anything changes.

O-rings for all four components cost approximately €100.
A PD element, probably costing around €350 or more, they're just trying to rip you off.

What do MWB 013, 018, and 023 say? Have you already logged them?

chli1976 wrote:

So, we are not arguing, we are just exchanging our experiences. Steffen probably sees it exactly the same way.

Yes.

If you buy the Bosch oxygen sensor (part number: 1417010997), it costs significantly less. Additionally, expansion bolts costing around 80 euros are needed if you're replacing all four.
It's not a minor intervention because the injectors have to be removed. That, in turn, should be done by someone who knows how and has the right tools for it. Otherwise, you might end up with even more problems. I wouldn't start working on it unless I'm sure that's actually where the problem lies, not just based on a hunch. That's what I mean. Pump nozzles are often better than their reputation suggests. I don't want to think about how many have already been replaced unnecessarily, just based on suspicion.

Please take the PD log with the engine running/warm.

If you want to check the AGR valve, simply create a log file. Don't buy a new one without proper diagnosis. You can always replace it later if your suspicions are confirmed. Similarly, you can test a gasket even if it doesn't appear to have any holes initially.
In this case, you perform a full-throttle test while driving in gear 3, 8, or 11. It's best to do this going uphill. The goal isn't speed; it's simply ensuring the accelerator pedal is fully depressed. This allows you to observe the air mass and boost pressure. A faulty EGR valve will be noticeable during this test.

We had an Audi in here last month where the EGR valve was permanently open.
"However, it started perfectly fine. But it was emitting a lot of black smoke. That's because it's an EGR valve with electric actuation. Those types often have plastic gears that can sometimes break. Your car doesn't have that issue; you have a pneumatic EGR system without any gears."

Here's a picture of the broken AGR valve for illustrative purposes. On the one on the left, the cover is crooked and can no longer close properly.

Okay, my dear friends,

Okay, so the EGR seems to be working correctly. The boost pressure and air mass are both within the specified values.
I had parked the car overnight with the front end facing downwards. Yes, unlike what I thought, it started right away. I will park it the same way today, as the vehicle is needed tomorrow.
Tomorrow evening, I'm going to do the test with the snout facing upwards, and I'm really looking forward to seeing the results.

Okay, I've created a log now; hopefully, I did it correctly. The car was warmed up properly, and all the temperature data seems reasonable. I started the log during the drive. At the very end, the vehicle was stationary, and the accelerator pedal was briefly pressed.
The LOG LOG file was created with a cold engine and while the vehicle was stationary.

Best regards,
Florian