W220 no boost after 5 min

[Thomas88]

Hey, I have a 2002 s320 cdi, w220 automatic. The car starts fine, idles fine, revs high and everything seems in order, however, after driving a few minutes (1-5min) the car looses all boost pressure.

I have icarsoft mb v2.0 where i can retrieve some data, such as maf sensor values, and inlet pressure.

The maf sensor seems to be working as the value increases with revs, and also the inlet pressure rises when revving. After a few minutes, the turbo disengages, and inlet pressure drops to 95kpa.

I have replaced exhaust manifold gaskets, intercooler, pipes to and from intercooler as these were leaking upon purchase of the vehicle.

There are no fault codes relating to vacuum or engine as i can find. the car is able to run through all gears and keep up with traffic, but overtaking is impossible.
The previous owner has disconnected the egr. There are no fault codes relating to this.
and also, the turbo disengages even if only idling.

So what could be causing this?

 

[LostKiwi]

The turbo is a load sensitive device. No load = no boost so I wouldn't expect it to do anything at idle.
It sounds like it dropping into limp mode. Any codes in transmission - that can also cause limp mode.

 

[umblecumbuz]

When this occurs, what are the maximum revs the engine can make? If it will not rev beyond, say, 3000, then this adds to LK's comment that the ecu has put the engine into limp mode.
Limp mode is not necessarily engine related - it is often caused by a transmission fault.
Puzzling though, because your iCarsoft unit should be able to isolate any transmission fault codes.

 

[LostKiwi]

Puzzling though, because your iCarsoft unit should be able to isolate any transmission fault codes.

Only if you look for them...

 

[Thomas88]

Thanks for replying
It revs all the way to redline, have checked gearbox and no fault on it, when i say turbo deactivates when idling, i mean that it doesnt reactivate after that. So even when only idling for 5 min, it will deactivate and not activate again when taking off. Restarting the car makes turbo activate again for a few minutes.
The engine functions as a completely normal engine without turbo, so it is like it doesnt know the turbo doesnt work, maybe. I didnt check what pressure it was able to make on max load, but that could eliminate leaking pipes and such.

 

[Botus]

Did this low performance issue come after you first "looked" at the car with your scanner. Whilst the EGR can be removed you have to bodge up the car to accept a bits missing. Not fully up to speed with how they do this (its bad form and just produces more cancer for everyone on the planet, so never read up on it), but I believe you have to mess with the engine management to get the car to cope with the modification. I'm wondering if the bodge is within the adaption area of the engine map and you killed it clearing faults. In which case the fix could be put the EGR back or get the ECU fixed with the software bodge. If that makes no sense, I'd move my next ramblings

I wrote these other paragraphs below before fully reflecting on your original post (hence edit above)

your terminology isn't correct, a turbo can't deactivate, you might not be getting boost when you expect or all boost you expect. but it doesn't have an on off switch. What I take from your explanation is the boost appears to drop off after 5 minutes of driving. If it revs out cleanly but performance drops off, do you think the overall condition of the engine is OK? is it regularly serviced - filters (air, oil, fuel) all often replaced and oil, coolant in good condition?

I'd check your code reader has all the updates it can (as they add features that may not be there when you first buy it). Then do a full scan and come back with what its saying.

Whilst many here frown on clearing then driving and testing again... most cars have a long list of old / random junk they pick up from never having been cleared after a fault was resolved years earlier in its life, and or the odd glitch where something as simple as voltage was a bit low. Thus car's often have a zillion codes of totally meaningless junk that can lead you astray. If it drives with no lights its fairly safe to wipe clean.... turn it off, start it up and go for 5 mile drive and check what things might really be a current fault.

However I caution against the magic one touch fix it button some code readers have, they should never be used. Wipe each one individually and if keen take a photo before clearing. One touch fix can wipe out individual "enhancements" the car has been adding to during its life to "adapt" to odd manufacturing issues with your car when it was built, and or to compensate for wear and tear during its life time.

FYI you should never drive a car hard or use more than 1/3 throttle till you've got 8 mile done to get thing moderately close to operating temperature

 

[oigle]

I agree with Botus. Most likely the fault is with your egr. Either fix the shunt, assuming it has one, or reinstall the egr to working condition.

 

[mioba]

Code readers are only as good as its user. get the car to a merc indy who knows what he is doing. this will end uo with misdiags everywhere and/or overly expensive

 

[onefortheroad]

What live data does the exhaust back pressure sensor give you at tick over. ?

 

[mersum1es]

My vote goes too for insufficient knowledge of EGR disabling...

 

[Thomas88]

Hi, i'll try to clarify,
The car was bought three months ago with a leaking intercooler, exhaust gasket, and the issue with the loss of boost after 5 minutes. As to service history, the car has probably been neglected. I have ordered new filters all around the car, to provide some needed comfort.

The boost issue did not change when replacing the intercooler, gaskets to turbo, and gasket to manifold. The turbo was replaced in 2010 by a previous owner, and it is a garret m4. Since egr was disconnected by a previous owner, i haven't had the chance to check what exactly has been done to accomplish this, i can mention wood screws in the old intercooler as part of the maintanance done to the car, wich obviously leaked, so i imagine that not everything has been done correctly when disabling the egr, or other things.

I'm away from the car this weekend, but will check for latest update on the icarsoft when i get back, and check the exhaust back pressure.

Had the car in for star diagnosis, at a mercedes workshop, and no fault on the turbo or egr system was discovered there either, the only thing on the engine was the maf sensor showing possibility of fault, but this was when the intercooler was leaking, and the workshop told me it most likely was caused by the volume of air passing through was greater then expected. But made perfect sense because of all the leaks.

 

[mersum1es]

EGR fault may also trigger MAF plausibility code since reading is not changed as expected when EGR is opened/closed.

When you get live data, read also temperatures...

 

[Thomas88]

So here is an update/solved status:
I reconnected the egr, nothing changed, took the egr apart and found a blocking plate installed by the previous owner. This blocking plate hindered the egr from opening. Removed the blocking plate and cleaned out the egr. Reconected it and it is now working fine. So thank you for your advices.

If there is anyone else with the same issue:
The maf sensor showed around 20 g/s egr when the egr was blocked when idling. When the egr is working the measurement dropped to around 12 g/s.
For my car the symptoms were black smoke if pushed past 2000 rpm, smell of unburnt diesel when idling, loss off power, loss of boost, turbo actuator disengaging after a few minutes (you can visible see the arm moving and you hear a distinct change in sound from engine when this happens during idling), inlet pressure drops to 95 kpa, no fault when reading codes and no limp mode or check engine light, not even with star diagnosis at authorised mercedes garage, engine had full rpm range.
Engine is 613.026 320 cdi 2002.

 

[mioba]

Well done to the OP. You must be well chuffed after your efforts.
Car certainly had a fee bodge jobs and pleased you have kept the car on the road.

 

[mersum1es]

when EGR is open, fresh air flowing into engine is smaller because its driving sh*t from exhaust side. Too low air ratio cause black smoke. EGR should close when revving up. I still believe there is something about EGR control or boost lines. If it 'disengages' turbo, that is a fault OR if that triggers no event or fault code it menas that car thinks everything is ok --> something in readings change so that turbo is needed to fold down. Are intake swirl flaps working? ECU not remapped? Boost lines double checked for splits or leaks (intercooler too)? What is the boost temp showing? Boost pressure when flooring the pedal?

 

[Thomas88]

Mersum1es, after the egr got sorted, the boost is there. Checked all vacuum hoses, actuators, solenoids. Everything on the vacuum side is fine, i havent checked the swirl flaps, the turbo actuator most likely disengages since the airflow on idle is greater than the expected value. Why this does not trigger a fault code other than maf i dont know. Intercooler and hoses are brand new, gaskets from exhaust manifolds and turbo all the way to egr mixing chamber are new. And i have no clue what has been done to the ecu since i bought the car running like this, previous owner says he repainted it and nothing else.
Since the egr was blocked, and this was not mapped in the ecu, i assume that the ecu is running original software. The black smoke could also come from too much diesel, since 20g/s is almost twice the normal idle value, i assume fuel trims are up way high, when disconnecting the maf sensor, the maf falls to 8 g/s and there is no smoke anymore since fuel trims down again. Maybe the smoke is blueish, but its a lot of it, and its hard to tell when driving.

 

[Botus]

thanks for coming back with how you got on...

seems like much much longer ago