Turbo Shrill Noise

[psmart]

The famous words have come back to haunt me 'sorted'! Further to my post on ML270 lockdown, the car is behaving, but since a heavy storm Thursday night, their is a high pitched shrill noise when the Turbo is coming on song, around 2-3000rpm. If you boot the car, the noise isnt their, if your cruise gently, the noise isnt their.

The noise resembles an amplified insect and is coming from the area around the Air-Filter, Boost Pressure Transducer etc. Cant pin-point it and the CarSoft software I have is useless if it cant read a fault code !!!

As the Boost Pressure Transducer registered a fault 3 weeks ago, but hasnt since, and no lockdown has occurred, has anyone else encountered this problem? It could be the noise of a leaking rubber gasket, or something more sinister. The car runs A1 fine, so cant fathom it!

 

[psmart]

Please refer to thread http://forums.mercedesclub.org.uk/showthread.php?t=16879&page=2 where Im continuing the saga.

 

[psmart]

In search of shrill......

...loud insect noise still emanating from Exhaust/Turbo system, even though a new Primary Cat to Exhaust seal installed and spent 200 quid getting a Turbo to Exhaust Manifold gasket installed evil: Im a Yorkshireman, and pointlessly wasting more than tupence fires the blood).

NB. On ML 270's, Mercedes decided that it would not fit gaskets between the Turbo and Exhaust Manifold, but it fits these Gaskets on all other 270's!!!!!

Anyway, after picking the car up after the above Gasket was fitted, driving to Liverpool and back and the noise is still there and load, decided to get down and dirty, take the engine bay apart... Pictures attached show a 270 engine stripped to get to the Turbo...

Basically, using stephoscopes, listenend to all possible junctions, but could not locate noise. If you know of the Turbo whistle, then imagine the peak of the whistle being replaced with an amplified cricket noise... from my deductions, as the Turbo vanes engage into the hot exhaust stream, they cause a larger backpressure between the Turbo and Engine, and there must be a leak (pin prick) or a sharp edge head on into the gas stream to cause this noise.

Anyone else ever come across this or similar problems? Engine performance is spot on, so no loss of pressure to the Turbo, no bearing or other noises eminate from the Turbo/Cat.... what is it?

Tomorrow, Im going to wrap PTFC tape around every joint and look for a burn point!

 

[psmart]

....Shrill has been found......

Eureka... after hours of PTFE taping, modified bearing stethoscopes (converted to listen for sounds) and Gas Leak Detector fluid, the culprit has been found... see photos. Its the Turbo Van Actuator bearing!!! There is a leak of exhaust gas through the actuator, as can be seen by the bubble at tickover.

Summary:
The car had several failed glow plugs at -30'c, but needing the car, I cranked it until it started and as a result, what I thought was a backfire occurred several times, which visually appeared as a puff of black smoke. After approx 1000 miles, I had the car chip tuned (remapped) and a short while later, going from 125mph to 0 (a traffic jam), the Shrill started suddenly. After a further 1000 mile approx, I got down and analysed and found that the Primary Cat to Turbo seal was leaking and the engine bay was saturated with unburnt carbon particles. Either the Turbo is faulty, or the unburnt carbon particles have caused premature wear to the Turbo Vane Actuator bearings or the chip tune caused an issue, or a combination of all three. Basically, when the engine starts up, at idle, it applies full actuation to the Turbo Vanes, bringing the vanes into the exhaust gas stream. When accelerating, the actuator arm reaches a position of about 1/3rd open, when the shrill occurs, so basically, the bearings have gone oval at this position, albeit slight amounts of leakage are occuring at all positions.

New Turbo required, only a matter of deciding wether I should replace it now or in 20,000 miles as it has no effect on the cars performance and the only detrimental item is the heat shield is being perforated by the exhaust leakage from the actuator.

I would have removed the Turbo today, if it wasnt for the fact that a HGV saw fit to have a fight with my ML on Friday, when it was parked outside the indie (to have the exhaust manifold to Turbo gasket installed). The HGV lost but caused damage to the ML (first real damage in its 5.5 year life) requiring repair tomorrow! Once you have removed the airfilters etc (30 min job), you have full access to the Turbo, and other than frozen bolts, should be a straight forward jobs to remove.

 

[psmart]

The saga continues.....

Replaced the turbo (new one), new cat, seals etc, cleaned out the entire inlet tract from airfilter to EGR valve chamber, cleaned all sensors, replaced oil with fully synthetic and pre-primed turbo with this oil..... started up, everything perfect, so left running for 10 minutes to allow oil to circulate around turbo and engine..... then revved up engine, wolfwhistle starts, then crescendo's into a shriek..... aaaarrrggggghhhhhhhh!!!!!!!!!!!

Two days running tests, analysis, reasoning and no closer to solution, hence Im hoping someone out in Cyberspace can shed light!!!

Firstly, please ignore my earlier info about the turbo not spinning up when the BPCVT is disconnected, THIS IS WRONG, I based it upon what I'd managed to read about these VNT turbo's, but like everything in life, hands on experience is way more valuable than 'reading'. Basically, the turbo spins all the time the engine is running, only the VNT actuator is set to a pre-set minimum (which prevents turbo surge on high engine loads, low engine speed). Thus, if the BPCVT is disconnected, as the engine speed increases, the turbo increases, but only to a low speed and low psi compressor output.

Next, description of tests made:-
1) BPCVT disconnected, turbo actuator remains at minimum, a very slight whistle is heard, no shriek.
2) BPCVT connected, turbo actuator moves to fully open position, when engine is revved, whistle starts, then actuator starts to move towards the closed position, approx 1/3rd travel distance but the whistle goes into a shriek. Continue to rev the engine faster, shriek disappears.
3) Compressor inlet and outlet piping removed, BPCVT connected, exact same results as in 2) above, shriek is heard. Thus the shriek is not coming from charge air in the Intercooler, Inlet Manifold etc.
4) Re-connected all pipework, and using a donor car (A C220cdi, which shares EXACTLY the same sensors as the ML270cdi except it uses a GT16V turbo instead of GT22V turbo) replaced all sensors, BPCVT. Exactly same results as 2) above.
5) Compressor inlet and outlet piping removed, BPCVT disconnected, partially blocked the exhaust to simulate excessive exhaust backpressure. No shriek heard, indicating that there is no exhaust leak in the exhaust system and also that the backpressure is not causing the turbo to shriek.

We made several other tests, including playing around with the vacuum control system, but to no avail. We didnt have a vacuum gauge or boost pressure guage, so could not see these values.

Road trials.... here is the interesting part. The car is running the standard ECU map (not the remap I have), and with the new turbo, new cat, all sensors cleaned, inlet tract (including intercooler) cleaned, the car runs almost as quickly as the car used to run with the ECU remap!!!!! It has no hesitation, no dead spots, very responsive, and requires very little throttle manipulation at slow speeds because the car seems to have loads of grunt!!!!! You may be asking, why am I so concerned on this noise.... quite simply, a noise from a turbo could be CHOKE, SURGE etc, each of which could cause the turbo to self destruct!

Fault possibilities which we havent inspected:
1) Vacuum system is faulty - ie. it could be creating too large a vacuum, thus keeping the turbo speed too high, meaning the turbo is reaching or going above its design speed, and causing overboost.
2) ECU is faulty - its closed loop system of turbo actuation by the BPCVT, and measurement by the Boost Pressure Sensor, Charge Air Temperature monitor and Mass Airflow Sensor is not being acted upon.
3) Something wrong with the exhaust system, causing back-pressure which in turn causes the turbo to CHOKE, albeit the test in 5 above seems to negate this.

Anybody got any ideas or clues as to possible areas? More importantly, does anybody know what CHOKE, SURGE and overboost sound like, audibly? I've scanned the net etc, but have come across nothing which describes audible conditions!

Many thanks in advance.

 

[psmart]

No one got a clue....?

...there are many sites talking about turbo's on the web, but none actually provide real info, diagnostic wise, only hypothesis and theoretics. Is this another black art area where those who know chose not to say?

 

[psmart]

Some more info...

....ok, more diagnostic info to see if anyone twigs to the problem.

1) Car flies, 0-60 timed today in 9 seconds for a standard ML270, no changes, other than a screaming turbo!
2) Vacuum is at 0.8bar at the reservoir, 0.6 at the turbo actuator. Upon acceleration, vacuum briefly blips and stabilises. Actuator drops to 0.5 bar then down to 0.2 depending upon revs etc. Pressures are identical on a known working C220cdi, but the actuator pressure differs slightly, problably down to the program in the ECU.
3) Boost Pressure up to 0.5/0.6 bar at standstill, so isnt overboosting.
4) Restricting the actuator arm by 1-2mm STOPS the noise completely, but initial acceleration suffers, but once on the move, no difference noticeable.
5) Blocking the exhaust and revving hard makes the noise appear lower down in the rev range and louder.
6) Disabling the actuator, blocking the exhaust and revving hard - no noise at all!

The Turbo is not making any mechanical noise and when in 4 above, is virtually silent, except for a whoosh of air as the compressor spins up.

So, looks like Ill have to strip the car down again and take out the exhaust - looks like a potential exhaust 3-way cat or muffler problem.

Anyone any comments to the contrary?

 

[psmart]

Mercedes Bodgemanship

I have found the first example of poor production quality on my car, and likely this problem exists on other ML's built around the same time as mine (MY2001).

To quote Sherlock Holmes, "When you have eliminated the impossible, whatever remains, however improbable, must be the truth", and so it has been finding this fault, which has been many faults masquerading as 1, caused by a singular build fault.

I'll refresh on the situation, clues and faults:
1) Glow plugs failed, car expected to start at -30'c. Witnessed puffs of black smoke seemingly coming out of the air inlet silencer or vicinity whilst cranking and backfiring. On advisement, added petrol to diesel, cured starting problem until Glow plugs replaced!
2) Car had the ECU re-mapped to gain extra power.
3) Car locked down at 120mph to 85mph, and after reading the fault code, the Boost Pressure Control Valve Transducer (P1470) was at fault.
4) On a high speed journey (125mph) (in Germany, so is legal!), car went from 125mph to 0mph in thunderous rain and stood in a traffic jam for 30 minutes. A screach or shrill sound appeared shortly thereafter, which occurred in line with the turbo whistle, as the turbo reached the climax of its whistle, the shreak or shrill would be heard.
5) Initial analysis showed that the Turbo to pre-catalyst exhaust junction was leaking. This was replaced by Indie.
6) The shreak or shrill still occurred, so Indie evaluated it as being the manifold to turbo junction, which on an ML has no gasket. Indie replaced this gasket. Indie also found the air-filter was completely blocked with carbon, so it was replaced.
7) The shreak or shrill still occurred, so I began getting involved, stripped down parts of the engine and began fault finding with various techniques, using chalk powder, plumbing tape, gas leak detector etc. Located the shreak or shrill noise coming from the Turbo VNT Actuator arm, which was haemorrhaging exhaust gasses. No other leaks.
8) Replaced the Turbocharger with a new one (you cannot get re-conditioned units other than from Garrett/Mercedes and they cost almost as much as a new one), replaced the pre-catalyst, seals, clamps, gaskets, oil feed lines. Cleaned out the entire air inlet tract, intercooler etc. On re-starting the engine after rebuild and usual turbo initial startup precautions, the shriek or shrill was still found to exist.
9) Disconnecting the complete air inlet tract (turbo compressor side), the shriek or shrill still occured.
10) Various tests were performed as above, to eliminate all possibilities with the Turbo control system. Found that a 2mm restriction on the VNT actuator arm would stop the shriek or shrill.
11) Road testing of the car showed a marked improvement in performance, even though the shriek or shrill was still occurring. The shriek or shrill was still coming from the Turbocharger and thus analysis, study, hypothesis and possibilities were worked upon, and as no leak could be found, the last vestige of possibility lay in the exhaust system, possibly causing a turbo CHOKE condition due to a restriction, or a sharp leading edge causing a noise which re-sounded back up into the turbo. Removing the exhaust system disproved this theory.
12) As per Sherlock Holmes, what was left, had to be the culpritt, and what was left was the engine valves, exhaust ports, and manifold. Previous tests had shown no leaks in the manifold or gasket, but as it was the only last vestige of possibility, we fabricated various stethoscope instruments and only by shear accident, did we locate a pin prick leak in Cylinder 1 at the nut! You could not hear this leak unless the turbo was spinning up at shreak point (which only lasted a short time as the ECU always backed down on turbo boost) and you had to be directly ontop of the leak with the stethoscope to hear it, otherwise the sound seemed to come from the turbo turbine! For people that dont know, VNT turbo's create approx 30% more backpressure than a standard wastegated turbo, and it was only due to this extreme backpressure as the turbo was spun up, that the leak occurred.

The complete exhaust system, turbo etc was then stripped down and analysis made, and the fault became glaringly obvious! Let me explain:
On re-assembly of the new turbo, it was noted that the turbo support bracket had worn from round to oval. I bought a new bracket and upon attempting to fit it, noticed that the bolt hole was more than 1 bolt width out of true. Turbo's are supposed to be mounted WITHOUT STRESS, and clearly during manufacture, the original turbo, bracket etc did not line up. We filed out the new bracket so that the turbo mounted onto the bracket, free of stress.
If you havent already guessed, then what has been happening is that the turbo turbine has been 'twisting' due to the stress of the bracket being over 1cm out of true. This in turn damaged the turbo to manifold seal, it more than probably damaged the pre-cat to turbo exhaust clamp and the final proof is that the cylinder bolt on cylinder 1 was also wearing away at the manifold. As no one has heard of exhaust gasses leaking out of the VNT actuator, it looks like this also could be caused due to the stress on the turbine housing, the temperatures it attained and the vibrations!! There have been many red-herrings in this search to find the fault!

If you own an MY2001 ML270, you may well encounter the above scenario, hopefully this thread will help you in its resolution either directly or passing it on to your indie.

I wonder if Mercedes would honour a warranty claim, given we have concrete proof (I will post photos in due course) and that a lot of the analysis and fault finding was done by a 30year experienced Chief university engineer, ie. a man of proven technical experience and standing!

 

[psmart]

Pictures

Some pictures showing the bracket in question and how we had to file the new bracket out to fit so that the turbo mounted free from stress and strain.

Also, with the new manifold, the whole assembly fitted closer and more snug than the components we replaced, just going to show that Mercedes do bodge production jobs!

 

[psmart]

To: W163ML270

Further to your email, I cannot respond to you as the forum blocks the options, so next best option is to reply on the thread in question.

'Hi I was just reading an old posting in 2006 about the shrill noise coming from turbo being under load, I have exactly the same issue and just had my turbo rebuilt and the noise was temporarily gone and has since returned. I just wanted to know the exact instructions to give to my technician to rectify this problem which has already cost me thousands of dollars. Any help you could give me would be greatly appreciated. Thanks.'

In my case, the shrill noise was caused by a micro crack on the exhaust manifold, only detectible by creating a stethoscope and listening around the manifold. The cause was because of a badly fitting manifold, which was twisting the entire exhaust/turbo attachments. When the turbo came on load, ie. it was creating back pressure onto the exhaust manifold because the turbo vanes were engaged in the exhaust gas stream, the pressure caused exhaust gasses to leak out of the micro crack in syncro with the turbo noise, so it appeared as though it was the turbo making the noise.

Hope this helps you.

 

[munga]

I think I have this problem too.

I recently had my intake manifold replaced along with Engine mounts, Front wheel bearing, Suspension bushing.

A new turbo was fitted by Mercedes about 70k kms ago.

I have a 2002 ML 270CDI and it has 176k km on the clock with full service history.

Does this mean the Exhast manifold must be replaced along with this bracket you mention or can the manifold be repaired?

Is the turbo okay?

Is there any sure way to say this is exactly the problem?

My car makes a screeching sound almost like a slipping belt (along with the tensioner) although I have just repalced this to make sure it is not. The sound only appears between 2000 and 3000rpm and usually under hard acceleration at low speeds and sounds as if it is coming from the general area of the turbo.



Any help would be very much appreciated.

 

[Mr.Lohja]

I think I have this problem too.

I recently had my intake manifold replaced along with Engine mounts, Front wheel bearing, Suspension bushing.

A new turbo was fitted by Mercedes about 70k kms ago.

I have a 2002 ML 270CDI and it has 176k km on the clock with full service history.

Does this mean the Exhast manifold must be replaced along with this bracket you mention or can the manifold be repaired?

Is the turbo okay?

Is there any sure way to say this is exactly the problem?

My car makes a screeching sound almost like a slipping belt (along with the tensioner) although I have just repalced this to make sure it is not. The sound only appears between 2000 and 3000rpm and usually under hard acceleration at low speeds and sounds as if it is coming from the general area of the turbo.



Any help would be very much appreciated.

If you did manage to sort this problem? Can you let me know what was it.? I have exactly same noise with same characteristics. Thanks in advance

 

[Gazwould]

1 decade