TDI Engine: Hard Starting After Idle Time

Hello to all diesel and petrol enthusiasts!

After I've already managed to solve my cold start problem (more on that later), I still need some helpful suggestions regarding a different issue.

I have a Volvo 850 TDI, manufactured in 1996, with 301,000 km on the odometer. It has an Audi AEL 5-cylinder engine and a VP37 ESP injection pump. The car has been converted to run on rapeseed oil (RME) using a dual-tank system, which has been in operation for 100,000 km.

'In the past, when starting the engine, it would immediately start up; drivers of TDI vehicles with healthy engines know what I'm talking about.'
Currently, if things go well, it takes 1.5 to 3 seconds for the engine to start. If the engine has been idle for a longer period, the time it takes to start also increases.
Yesterday, for example, it was so bad after being idle for two days that I thought I wouldn't be able to start it at all. It took over 10 seconds before it finally started, and then very slowly.
The shorter the time the engine was off, the easier it starts again. But once it's running, it runs well – full power and no hesitation or anything like that.

The injectors have been replaced. The fuel injection pump belt was also recently renewed, and the valve timing has been adjusted to the earliest possible point within the allowed tolerance.

There is no air in the system; this can be seen through the transparent tubes of the short-circuited return line (due to the installation of a low-pressure drop device).
The engine temperature sensor and the fuel temperature sensor of the ESP are working correctly.
The starter motor is turning over quite freely, and the battery is only about 6 months old.
The spark plugs are new, but the glow plug system shouldn't be a factor at the current temperatures.

'Due to additional cold start problems that the engine was still experiencing recently, I removed and cleaned the electric shut-off valve of the ESP system. The problem was that the engine, once warmed up, would no longer restart in warm weather (above >25°C). Cooling down periods of 1.5 to 2 hours were necessary. This behavior has disappeared since I cleaned the valve. It seems that the valve was sticking when it got hot and preventing fuel from being injected.'

Could the valve also be responsible for the poor cold start performance?
Does the valve or something else within the ESP fail to maintain pressure after it's shut off, causing the pressure before the injectors to gradually decrease and then requiring a longer cranking period during startup to rebuild that pressure?
Is it actually true that the pressure in the high-pressure lines leading to the injectors is maintained after the engine is switched off, or does it drop back down? If the pressure is maintained, is this due to the valve on the electronic stability program (ESP), or some other device???
Or am I on the wrong track?
The ESP itself gives a dry impression externally...

It might also be worth mentioning that if the engine starts with difficulty (cold start) and is then immediately turned off and restarted right away, it starts instantly without any delay.
If you slightly open the fuel lines at the injector holder combinations and turn the ignition key, the fuel will initially drip out slowly, without pressure. Only when the engine is almost starting does the fuel spray out of the line under pressure.


Best regards.
Morti.

Hi,
'Well, to me, it sounds very similar to what I experienced. Although, I didn't just have a problem for 10 seconds, and it really seemed like the car was about to catch fire.'
The suspicion that there was obviously air in the ESP (Electric Power Steering) has been confirmed for me. After working on the pump, I bled it through the return line, and sure enough, a lot of bubbles came out.
I don't know if PÖL is as aggressive as biodiesel, but maybe you have a tiny puncture mark in a seal somewhere else on the ESP.
'Bleed the power steering system after the car has been sitting for a long time; then you'll see if you have any air bubbles and know what's going on! It doesn't take long and it costs nothing!'
Best regards,
XOXO.

Hmmm...

To me, this sounds like it could be related to air intake. The fact that the starting problems are getting worse over time supports this. Seals, especially the crankshaft seal on the drive shaft of the ESP (Electronic Stability Program), are likely culprits in this case and relatively easy to replace. However, you claim not to 'see' any air leaks. It could also be a vacuum leak. Another possibility is a broken spring in the high-pressure section of the ESP. This is something that's often reported. The symptoms are similar to what you're experiencing: gradually worsening cold starting behavior combined with increasingly difficult warm starts. Alternatively, it could be a failing rotor vane pump within the ESP.

Here's the translation:

'How is your boiler conversion set up? Is it a short-circuit return system? Which heat exchanger are you using, and where is it located? Are there any leaky switching valves that might be drawing in air, even if they aren't letting fuel out? Where is the pre-feed pump located, and what type is it? What pressure does the pre-feed pump generate in the fuel circuit? (Bosch specifies a maximum of 0.2 bar here. I think 0.4 bar shouldn't be a problem. More than that could potentially dislodge the aforementioned shaft seal from its seat and promote air entry.)'

To pinpoint the air-related issue, it might be possible to temporarily install a pre-lift pump very close to the fuel line near the ESP and, if feasible, power it on with voltage when the ignition is engaged. If this works and the engine starts easily, then it's likely an air problem.

Hello Kussi and Lappen, thank you both for your contributions.

Okay, so if there's air in the fuel system, you can usually see it clearly through the transparent hoses – air must be excluded from the fuel lines. I think the fuel system is tight, and I've also checked all the hose clamps, etc., again.

Is it possible for air to enter the ESP (Electronic Stability Program) system through a different route, which would not be visible in the fuel lines but would only reach the high-pressure lines leading to the injectors?

Today, I had some air in the system because I replaced the filter for the diesel tank in the spare tire well. It's just a very small one: 'http://www.monopoel.de/shop/s_filter-812.html.' There wasn't much air at all, and initially, it still started. After a short period of inactivity, however, there was nothing more to be done. Then, the newly purchased ball pump came into use: 'http://www.monopoel.de/shop/s_hand-ball8.html'. I opened the return line to the main tank and got the air out that way.

But the original problem still exists, even if there is no visible air in the system.

@Lappen: My PÖL conversion is based on a 2-tank system. In the spare tire well, there's a 33-liter tank containing diesel fuel, which is used for warming up and flushing the ESP towards the end of the journey. The vegetable oil is in the main tank. Before it reaches the ESP, it is brought to 'http://www.monopoel.de/shop/s_wt-a8.html)' using an A8 heat exchanger, bringing it up to coolant temperature.
In a closed-loop system, an electric auxiliary heater can be activated as needed.
There is no pre-lift pump; fuel is only drawn in by the electric fuel pump (ESP).

Could it be that there's simply not enough air being circulated for it to be visible in the transparent fuel lines? Or perhaps steam bubbles are forming due to the fuel being heated to 70°C? But if that were the case, it should have happened before as well, and I didn't experience this problem then...

What's the cause of this feather break in the high-pressure section? What exactly is happening there, and what are the consequences? How can I determine if the pump is affected by this?

I need a little more help understanding the processes in high-pressure fuel lines. Do they typically remain under pressure even when the engine is turned off? Or is there simply fuel inside without any pressure?
Is the shut-off valve of the ESP responsible for ensuring that fuel remains in the lines? But even without the valve blocking it, shouldn't fuel be unable to flow back into the ESP? In that case, air would somehow have to enter through the injectors... is that conceivable? Perhaps through the oil return lines?

Thank you very much in advance...

Regards,
Morti.

Hmmm...

Okay, to me, this looks like air is getting in. What are the transparent hoses connected to? Just the supply line, or also the return line? The supply line alone doesn't necessarily tell you much. If you have a transparent hose on the return line as well, what does it look like shortly after starting up? Is it foamy? Are there bubbles? I'm almost betting that's the case... If air is actually being drawn into the system, then the wing rotor pump could be running empty. Consequently, it will take a moment for it to refill and build up its operating pressure. Only when this pressure is reached will sufficient pressure be available for the high-pressure (HD) component (which generates several hundred bar from a pre-pressure of about 5-6 bar). Only then can the HD lines be pressurized according to the firing sequence, allowing the ESD valves to open and inject fuel. The HD lines are only under pressure during the injection moment.

Okay, let's try something else: If you loosen the locking nuts on the injectors and someone else starts the engine, does diesel immediately flow to the injectors? Or does it take some time?
If you hang a bottle, say 1 liter of diesel, on the hood (as a makeshift fuel source) and connect the hose from that bottle directly to the ESP (Electronic Stability Program), will the car start immediately, or will it take some time?
How old are the injectors? What is the spray pattern like?

I'm assuming that you rinsed it properly and for a sufficient amount of time before setting it aside!

Hello Lappen,

You are right about the assumption . The ESP has pure diesel available for starting. By looking at the color in the transparent hose, you can always clearly see whether it is pure rapeseed oil (RME), diesel, or a mixture of rapeseed oil and diesel. In city traffic, I only drive on diesel anyway, because the rapeseed oil does not burn optimally enough in the partial load range, which can lead to rapeseed oil entering the engine oil. I only switch to rapeseed oil for longer journeys on country roads or highways.

'What are the transparent hoses for: only the supply line, or also the return line?'

'Due to the short circuit loop, the fuel that is in the return line of the ESP automatically returns to the supply line after passing through the switching valves, where there is approximately 30cm of transparent tubing. Therefore, nothing flows back into the tank, which would be the case without a PÖL modification. This means that if there is air in the return line, you can see it again in the supply line shortly afterwards. The short circuit loop configuration is necessary to keep the fuel at temperature when operating with a PÖL system. Otherwise, it would flow back into the main tank and cool down again.'

'If you also have a clear hose in the return line, what does the image look like shortly after starting? Foamy? Bubbles? I'm almost willing to bet...'

Let's say, how about a case of beer?
No, really, there's nothing visible that would suggest the presence of air. If there is air in the system (like yesterday when I replaced the small fuel filter), you can immediately see it. Then you'll notice that everything becomes foamy, or that bubbles rise up when the engine is turned off.

- 'If there is actually a leak in the ESP (Electric Submersible Pump), then the impeller cell pump may run dry. Consequently, it will take some time for the pump to refill and build up its operating pressure again.'

If the wing cell pump were empty, wouldn't you be able to see air?

-'Only when this pressure is reached will the high-pressure system (which generates several hundred bar from a pre-pressure of approximately 5-6 bar) become available to the ESP. Only then will the high-pressure lines be pressurized according to the ignition sequence, and the injectors will open and inject fuel. The high-pressure lines are only under pressure during the injection moment.'

Let's assume that the wing cell pump isn't running dry, but is somehow weak and no longer able to build up the full initial pressure. Could this also be causing the starting problem?

Is there any residual pressure remaining in the ESP (Electric Power Steering) system or high-pressure lines after the engine is turned off? Or is all the pressure completely released?

'Okay, then let's try something else: If you loosen the retaining nuts on the injectors and someone else starts the engine, does diesel immediately flow to the injectors? Or does it take some time?'

'The diesel fuel doesn't immediately spray out when attempting to start the engine. Instead, it only trickles out hesitantly at first. Only when the engine is almost starting does one notice that there is significant pressure, and the diesel fuel sprays out from the fitting.'
I haven't tried the 'Falldiesel' method yet, thanks for the tip.

- 'How old are the injectors? What is the spray pattern like?'

The injectors are new. Consequently, the spray pattern should be good.


Thanks, buddy, for your constructive questions.
I would be happy if one or two more people joined in.
I know it's a tricky case. But that's what makes it so much fun to solve, right?



Regards,
Morti.

OK!

If the system doesn't reach full pressure immediately when attempting to start, it suggests that the injection pressure of the ESP (Electronic Stability Program) is not sufficient to open the injectors. If you are certain that there is no air leak, then there are two other possibilities: either the vane pump is not working optimally, or the springs mentioned earlier may be broken. Check the technical articles; they describe the springs and how to replace them!

Try that with the 'Falldiesel' (likely a specific product or brand). I'd be interested to see how it works!

morti850tdi wrote:

So, the diesel doesn't immediately spray out during the initial start attempt. Instead, it only trickles out hesitantly. Only when the engine is almost starting does one notice that there is significant pressure involved, and the diesel sprays out from the fitting nut.

So, the pressure increases significantly with the speed.
One possible reason could be that the "centrifugally driven" impeller of the pre-pump is not sealing properly at first.

I'm not sure to what extent the sticky residue from old materials might be involved (just a thought).

If you have time, try giving it a jump start during the next predicted difficult start. Alternatively, connect a battery charger to the battery for about 30 minutes beforehand.
If that helps, then the problem is likely related to the starting speed, and you might want to consider a new starter motor and/or new cables between the starter motor and the battery -> see our article on troubleshooting starting problems.

Off-Topic:
Have you replaced any other hose sections with clear material besides the standard clear diesel line (before the pump)?
If so, I would be interested in the source of supply, because nothing is transparent anymore with PD, and the usual clear fuel lines become hard over time .

@ulf

In my opinion, if there's a problem with the starter motor and its associated resistance, the engine would also be harder to start when it's warm.

lappen wrote:

In my opinion, if there's a starter motor and transition resistance problem, the engine would also be harder to start when it's warm.

Basically, yes, but if the pump blades were easier to move and sealed better when submerged in warm liquid, while the substance itself caused them to extend more slowly after a period of inactivity (due to tendencies to become sticky or something like that)...?

Okay, please provide the German text you would like me to translate into English. I will only provide the translation.
@morti:
As an alternative to jump-starting, you could try warming up the low-pressure side of the pump thoroughly (using a hot air blower / caution: avoid melting any hoses or plastic parts!), because, as far as I know, any deposits or build-up become more elastic or dissolve with increasing temperature.

The pump is properly purged only when it still feels quite warm about 30 seconds after the hot air blower has been turned off. Therefore, allow yourself at least 10 minutes, or even more!

If it starts more easily than usual, that supports my theory about the presence of sticky oil residues.

Hi Ulf, great that you're joining in!

'Regarding the starter motor, I had also considered it, but I agree with Lapp's opinion that it would always make a very loud noise. However, once it starts, and you turn it off briefly and then restart it, it generally starts well. Not quite as quickly as before, when you could start it within a second of turning the key, but still pretty good.'

I've just discovered something new:
The blind plug on the leak oil line of nozzle holder combination No. 1 (on the passenger side) has a very small crack! The leaking fuel is hardly noticeable. However, I wrapped a clean handkerchief around the nozzle and drove around. Afterwards, there were traces of diesel on the handkerchief. Then I took a closer look and discovered the fine crack.
Could this be the cause of the error? I'm going to replace that part now and will report back later...

Regards,
Morti.

Off-Topic:
Have you replaced any other hose sections with clear material besides the standard clear diesel line (before the pump)?
If so, I would be interested in the source of supply, because nothing is transparent anymore with PD, and the usual clear fuel hoses become hard over time .

Yes, there's another clear diesel hose available. Please check 'www.monopoel.de' under the 'Products' section, then select 'Hoses,' then 'Fuel,' and finally browse the various polyamide hoses.

'Highly popular polyamide fuel line resistant to diesel, vegetable oil, and RME. This transparent line is temperature-resistant up to 90°C and offers excellent visibility. Ideal as a feed line for cars with very tight installation spaces, as it is vacuum-stable and highly flexible, and also allows for easy detection of potential air leaks in the fuel system.'

The transparent original part in front of the ESP is quite yellowed, which makes it difficult to see through on mine.

I just removed the blind plug and took a closer look. There was a tiny hole in it. I cleaned it and wrapped it with duct tape; it should be sealed now. I put the plug back on. Tried to start it... 'organ' sound... 'organ' sound... 'organ' sound... but it won't start.
Checked: Now I could see a lot of air escaping, just by opening the plug...
So, I used the air pump to release the air again.
The next attempt to start the engine was difficult, but it eventually started.

I also noticed something else: after about 4 seconds of turning on the ignition, there's a clicking sound coming from the ESP (Electronic Stability Program) unit. Could this be related to the shut-off valve, or is something else causing the click?
If the shut-off valve doesn't open until that late, wouldn't that mean the engine can't start before then?

I believe I can disprove Ulf's theory about sticky fuel injector components: once the cold engine has started, if you turn it off after a few seconds and then restart it immediately, it starts right away. Therefore, the pump is still cold, so it probably isn't related to any temperature issues...

Regards,
Morti.

morti850tdi wrote:

Check out "www.monopoel.de" under the products section, then go to hoses - fuel - and then look at the various polyamide hoses. . .

Thank you .

Quote:

I also noticed something else: there's a clicking sound coming from the ESP after about 4 seconds, after I turn on the ignition. Could it be the shut-off valve, or is something else clicking there?

I know this as deactivating the MSW so that unnecessary power is not consumed for too long.

Quote:

Once the engine has started, even if it's cold, and you turn it off after a few seconds and then start it again, it starts immediately. The pump is therefore still cold, so it shouldn't be anything related to heating...

You won't get away with that so easily.
If it's a sticking problem and the wings have become loose during playing, they will likely remain movable for some time.

But to clarify:

Quote:

So, the diesel doesn't immediately spray out when attempting to start. It only trickles hesitantly at first. Only when the engine almost starts does one notice that there is significant pressure, and the diesel sprays out from the fitting.

Does the diesel start spraying properly only when the engine speed increases due to the initial ignitions, or does it spray more forcefully sometime during the cranking process, and only afterward do the first ignitions occur?

I know this as deactivating the MSW so that not too much electricity is wasted for a long time.
So, if the MSW hasn't started after 4 seconds, is it shut down? You always learn something new...
I'm still not entirely comfortable with that shut-off valve; I wouldn't want it to get stuck again.

Does the diesel engine start spraying properly only when the engine speed increases due to the initial ignitions, or does it spray more forcefully at some point during cranking, and then only after that do the first ignitions occur?
I think he's already spraying quite a bit while playing the organ, and then the first signs of ignition appear. But I'll investigate that more closely.

But tell me, how would a small leak in the oil lines (which wordplay ) become noticeable? Would air also be drawn in, or would only a little fuel leak out?

morti850tdi wrote:

But tell me, how would a small leak in the oil lines manifest itself? (That's quite a pun, ) Would air also be drawn in, or would only a little fuel leak out?

I suspect that while the engine is running, some diesel fuel leaks out (see tissue), but when the engine is off, the diesel fuel drains back and slowly draws air into the system through the crack.
The extent to which the pump can run dry likely depends on the piping configuration and the channels within the pump itself.