Turbo Loss When Engine Cold

[dt1]

Thanks, that explanation was helpful. I found and removed the filter but it did not help.

[seatkid]

check the pipe going into the vacuum reservoir, this is a small globe located at the front of the engine about halfway down. Could be loose.

 

This post has some information about the vacuum system, however a number of images have disappear because the people who contributed them have pulled them. :D

[NikpV]

This post has some information about the vacuum system, however a number of images have disappear because the people who contributed them have pulled them. :angry:

 

I'm sure those images were there earlier today ;)

[mumof4]

Nope..they werent there im afraid...as it was the same link Neil gave me yesterday and the pics were missing then.

[NikpV]

Nope..they werent there im afraid...as it was the same link Neil gave me yesterday and the pics were missing then.

 

 

must have been a while since I last looked at that thread then

[Smilge]

;) Clutch ........????? :angry:

[dt1]

I've not been able to check all the pipes yet; still trying to work out which they are! Pity some of the photos are missing from the thread mentioned. However, thanks for drawing my attention to the vacuum reservoir. The tube to it was quite badly chaffed and would have failed before long. However, when I cut it to bypass the bad bit it hissed so I assume this meant the pipe and reservoir were still holding vacuum. I don't suppose this means the whole vacuum system is OK (wishful thinking)?

 

If I want to replace any vacuum pipes do I need tubes specifically desinged to fit the car or just a 'generic' vaccum hose? The ID of the hose is about 4.1mm so does one use anything of the same diameter?

David.

[tim-spam]

It is a very good idea to check / replace all of the vacuum pipes, as this is comparatively cheap and easy to do. However, it sounds as if your system is holding vacuum pretty well. In addition, vacuum leaks would tend to reduce the level of turbo boost - your problem is intermittant over-boosting. This could be caused by a blockage somewhere, so it is definitely worth checking around.

[scrote]

I have exactly the same problem - turbo cuts out once a day from cold and is fine there-after. I have done most of the checks as described above but no joy yet. I have decided to live with the problem for the time being as Ive hours pulling pipes off and cleaning them etc etc.

 

Have you any update ?

 

Scrote

[tim-spam]

Does it go into limp mode every time after a cold start, or can you avoid it by driving gently until the engine is warm?

 

To avoid this problem being caused by a sticking VNT, it is a good idea to regularly use revs up to and exceeding 3000rpm. This ensures that the VNT mechanism gets regular exercise - if you regularly change up at 2000 to 2500rpm, the mechanism will spend most of its time in the 'full boost' position.

[dt1]

I can consistently prevent mine from going into limp mode by keeping the revs below 2000 while cold, and consistently trigger it by accelerating hard when cold. The new MAF gave me more power so this has kept me going. I have not checked for a sticking VNT yet so thanks for making that connection as a possible cause of these symptoms. I seldom accelerate hard so its had over 5 years of similar treatment. I continue to have only the "17965 - Charge Pressure Control: Positive Deviation" error since I started monitoring this. I have also experienced limp mode once recently when accelerating hard while hot. I'll try to check the VNC this weekend.

David.

[tim-spam]

I seldom accelerate hard so its had over 5 years of similar treatment.

This could be a possible cause of the sticking VNT - it will be spending a lot of time in its maximum boost position, whilst there are sooty deposits accumulating on the mechanism.

I have heard that some people have improved / cured this problem by changing their driving habits to regularly use higher revs, changing up at around 2500-3000 rpm, even when cold, and using up to 4000rpm at least once a day.This forces the VNT mechanism to move through its full travel regularly, preventing the sooty deposits from building up sufficiently to cause the mechanism to stick.

[dt1]

I have now checked the VNT movement. It is very small between starting and stopping the engine. I have two sets of photos below to illustrate this.

 

There is a audible vacuum 'action' and obvious physical 'jolt' from the actuator when switching the engine on and off. So I think I have good vacuum but sticking vanes? I assume the vanes free up when the metal heats up and expands. Does this sound likely? Next steps?

[dt1]

I have now also tested it when hot. Much greater movement. The shaft is about 6mm diameter so the movement is about 8mm when hot and I would guess about 2mm when cold.

[Guest VR6!]

Give it a few good hard runs and see that helps clean out the crap as tim suggests! then see if the cold running improves? whats known in the trade as an Itallian tune up! :blink:

[El Dingo]

Give it a few good hard runs and see that helps clean out the crap as tim suggests! then see if the cold running improves? whats known in the trade as an Itallian tune up! :blink:

 

This used to work a treat on my mother's old car. You know the story; car driven only around town, never got revved. She used to ask me to 'give it a tune up' when it got a bit clogged up, so I used to go down to the dual carriageway and hold it in a low gear at high speed for a mile or so. Worked every time. :10:

 

Now she has a (VW based) Skoda with fuel injection - never has any problems... ^_^

[dt1]

Good idea. I will do this for a while and see if it helps.

David.

[tim-spam]

The problem would seem to be that the vanes are only sticking when the engine is cold. I would not recommend thrashing a cold engine, but ensuring that the VNT mechanism gets a good workout from time to time is certainly a very good idea.

 

As a next step, have you a method of applying a vacuum to the diaphragm from an external source? If so, you could try exercising the mechanism manually with the engine cold - I don't know how difficult this would be however, having never tried it myself. If it was possible, however, it may scrape away some of the deposits and help to free up the mechanism. By this, I mean manually forcing the rod to move through its full travel several times.

 

You may find the following interesting: www.technologie-entwicklung.de/Gasturbines/VNT15-Turbo/vnt15-turbo.html

 

Failing this, either cleaning your own turbo, or fitting a reconditioned unit is the only real solution. There are a few turbo reconditioners out there who will charge a lot less than a main agent for the parts.

[dt1]

I've been trying to get my mind around how the VNT works and how this could explain my problem. Thanks for the link and other advice Tim and others.

 

It seems that the vanes should normally be wide open when the engine is off (the no vacuum failsafe position), fully closed when the engine is first started (to maximise low rpm boost from the small exhaust flow), and opened again as rpm increases to manage the turbo speed/boost (as exhaust flow increases).

 

My photos suggest that I get stuck somewhere in the middle of this movement when the engine is cold. Exercising this movement with different rpms should free it up a bit and help confirm that this is all or part of my problem. I will drive with a wider rpm range initially to see if this helps.

 

How hard is it to remove the turbo and how practical is it to open it up to clean?

 

What I am now a little puzzled about is how limp mode works. It feels like and is described as 'no turbo boost' but the turbo speed is mechanical and it seems to me that this can only be influenced by the vane position. I would be interested in knowing what the ECU 'switches off' when it goes into limp mode.

David.

[seatkid]

The amount of vacuum applied to the vane actuator is regulated by the TCS - Turbo Control Solenoid, also known as the Boost Control Solenoid. This is mounted on the bulkhead behind a shroud to the rear of the Turbo. The ECU send a varying voltage to the solenoid to vary the vacuum to the actuator which in turn varies the boost pressure. When the ECU sees the control system is not responding correctly i.e. the boost pressure is either too high or too low - it simply switches the boost off - limp home mode. Switching off the ignition resets limp home mode.

I would not recomend you open up and clean the turbo - thats a specialist job, but you can try to lubricate the vane actuator shaft, and IIRC the vane vacuum actuator can be removed. Before you start stripping the Turbo off, double check all the vacuum circuit and check and clean the TCS which is known to give problems occasionally.

 

see next post...

Edited November 13, 2006 by seatkid

[tim-spam]

Limp mode also dictates reduced fuelling of the engine, and prevents the engine from reaching high revs under load - in effect, this will restrict the level of boost and prevent engine and turbocharger damage.

 

As for stripping the turbocharger, it looks fairly simple, but there is no mention in the article of any sealing rings or gaskets - perhaps there aren't any, but it would be good to know this in advance! By the way, before attempting to lubricate the VNT machanism, remember that the turbo gets very hot, and this may end up making the problem worse if the oil burns and leaves behind any hard deposits.

 

My car very occasionally suffers from this problem when very cold, so I am regularly changing up at around 3000rpm when cold in the hope that it will get no worse - we shall see....

 

An interesting thing is that, when it is about to switch into limp mode, there is considerably more engine vibration, which would seem to confirm that it is actually over-boosting, and not just suffering from inaccurate data from the t-map sensor.

 

This thread has provided a lot of interesting information, especially from dt1 - thanks.

[seatkid]

"17965 - Charge Pressure Control: Positive Deviation" indicates that the turbo boost pressure is too great - i.e. the vacuum is not beeing dumped quickly enough and/or the vane actuator is not returning to its default off position quickly enough. Often this is the fault of the TCS being dirty or the "atmospheric" connection to the TCS is blocked - there is a small "mini" air filter connected on the end of the "ATM" pipe - it can get blocked. A simple test would be to remove the "ATM" pipe on the TCS and see if its any better.

 

For reference here a few photos of where the TCS is and what is looks like on a MkII

 

Its behind a shroud on the rear bulkhead:-

 

Three pipes are connected -

VAC vacuum supply, ATM atmospheric connection,and OUT vane vacuum

 

Close up of a TCS, the ATM connection is the black one....

[scrote]

As per my previous entry, My bus has been suffering the same problem for the last year and i have tried changing/flushing/blowing/cleaning nearly every component under the bonnet. It gets a good run of 30 miles most days and I regulary rev it hard when warm and rarely get much smoke from the exhaust so Im happy theres no big carbon build up.

 

However, there is one thing that gives the dirty sticking vanes theory some cridibility. Last winter we took the bus fully loaded on a 500 mile round trip - When setting off in very very cold weather the turbo cut out at least three or four times in the first 10 miles it then behaved for the remainder of the journey. After that trip the turbo did not cut out again for another two weeks - and then it started again and has been doing it ever since.

 

There is one other thing that i have noticed with the bus and i dont know if dt1 has the same - but the egr valve leaks oil down the back of the engine - not a huge amount but could it be relevant ?

 

Scrote

[tim-spam]

Just heard of someone with a Golf TDI 130 who had the positive deviation fault code and the consequent tripping into limp mode on a regular basis. He traced the problem to a sticking VNT mechanism, and tried a product called 'Power Boost', which is sprayed into the intake and is supposed to clear carbon deposits from the intake and exhaust manifolds (including the turbo). Here's a quote:

 

"Powerboost definitely worked although I had to use one and a half cans due to the amount of carbon build up, costs around

[scrote]

Having already spent hours on the bus I will give it a try and will report in due course

 

all others wishing to give it a try maybe interested in cans available on e bay

 

type in the following with the spaces EKOTEK POWER BOOST - and you will find cans available for