Volkswagen Eos Forum banner

Overboost (P0234) on BMM engine

8K views 19 replies 4 participants last post by  aku-aku 
#1 · (Edited)
Last summer, when I was out driving, my Eos went into limp home mode, so I pulled over, checked everything, restarted the car and all was fine. Checked the fault memory, noticed a code for overboost, didn't think anything of it, and since it didn't recur, more-or-less forgot about it.

This summer, when I was doing about 120 on the motorway on a hot day (around 30C), it came back with a vengeance. Cutting the engine and restarting would remove limp home mode, but only briefly, and it would return to limp home mode after a short period of driving. After it happened three times, instead of the flashing glow plug light, the engine light came on and stayed on - although again, restarting the car would remove the flashing glow plug light and limp home mode. On the home journey later that day I took it easy and managed to avoid limp home mode.

Given that the problem seemed a bit more serious, it was time to do some research. Here's a summary of what I've learnt: these engines have a variable vane turbo, which means that the turbo has vanes that alter the air flow to change the amount of boost. The position of the vanes is controlled by a solenoid that creates a vacuum that moves an actuator. Several things could go wrong here: the various air hoses could leak; the air pressure sensors could go wrong; the solenoid could malfunction; or carbon can build up on the dirty side of the turbo. (The dirty side is where the exhaust gases flow through a turbine, which drives a fan on the clean side that pushes more air into the engine.) In the case of overboost, the most likely cause is that carbon buildup has caused the vanes to stick, and the turbo cannot respond to a decrease in turbo pressure quickly enough. And what is overboost? It's when the turbo gives more boost than the car wants it to. The engine computer decides how much boost it wants to deliver and adjusts the solenoid's position to change the amount of vacuum which moves the vanes. Later cars have the vanes electronically controlled rather than having the vacuum bit in the middle - I guess the vacuum-operated turbos can work entirely mechanically rather than having the computer controlling how much they suck. ;)

The turbo vanes can be tested using your diagnostic tool of choice: start the engine, then go into the engine module, basic settings, channel 011 - this will start a test that will spin the engine up to 1400rpm then move the vanes from minimum boost to full boost and back. Here's a video of what it should look and sound like. I checked the frequency of the whistling sound on that video, and the tone of my turbo is around 10% lower, which suggests that the vanes are not fully opening. Unscientific I know - the ideal would be to know what boost pressure numbers I should see - but all I have to go on there is that someone on the internet wrote that it should be around 150mbar and I get more like 100.

So, first step is to stick a bottle of Wynn's Turbo Cleaner in the tank, and after a few hundred km trying to keep the revs around 2000 to ensure decent exhaust gas temperature to help burn the carbon off, see if there's any difference in the measurement. If that doesn't work, sounds like the turbo needs cleaning.

Unless one of you has a better idea, of course...?
 
See less See more
#2 ·
Hi there

Not sure how much affect small bottle will have.
Do you have a garage nearby that has this type of service...... https://m.facebook.com/terracleanmotherwell/?locale2=en_GB

I've seen it in action on a petrol turbo engine. All sorts of stuff came out the exhaust while the 'fuel' was running through the system.

Just a thought.
 
#3 ·
That's definitely something to look into. I'd figured that, if off-the-shelf chemicals don't solve it, the next thing would be to take it to a mechanic who would disassemble and clean the turbo, but I'll have to see if there's anyone nearby that can offer Terraclean.

I am still skeptical about the power of a fuel additive, but they claim it can free sticking turbo vanes, so I figured it was an easy and cheap first step. Their advice to add it to every third tank of fuel seems a bit much though!
 
#4 ·
I have tried many of these chemical cleaners to cure the vanes sticking even heard of using oven cleaner. Not one of these quick fixes lasted more than a week when the fault would come back with a vengeance, the only permanent cure is either a strip and clean or a new/recon unit IMHO.

Mick
 
#5 ·
I have tried many of these chemical cleaners to cure the vanes sticking even heard of using oven cleaner. Not one of these quick fixes lasted more than a week when the fault would come back with a vengeance, the only permanent cure is either a strip and clean or a new/recon unit IMHO.
Yes, I'm prepared for that to be the outcome :) I'm guessing it pays dividends to follow voxmagna's tips on driving style to look after things by keeping the revs around 2k, but it's hard to imagine that putting something into the fuel would be so magic as to have the required effect on the dirty side of the turbo after having been through the engine, and there's definitely no substitute for cleaning the turbo out properly if that's the cause. I'm hoping that at least this should give me an indication of where the problem lies, as a substitute/shortcut for paying for mechanics to scratch their heads...
 
#6 ·
I think the long term solution is to look into 'catch cans' and find somewhere to fit one? I have some parts and silicone hoses, but haven't got around to doing it yet. As the engine gets older there will be more blow back carrying oil over which burns then gets recirculated around and around. Some oil on the input side of the turbo is good, but it can be too much.
 
#7 ·
Some unscientific results... after running through the tank with turbo cleaner in, the boost pressure test returns 938-1050mbar (lowest value with the valve closed to highest value with it open), where before it was 928-1020. Seems like it made a difference, but if the high and low values are supposed to be 150mb apart then the turbo still needs a clean - but I don't know if 150mb is correct. Could someone with a BMM engine please do me a favour and run the Basic Settings 011 test on the engine module and see what the numbers are, so I know if I'm aiming at the right place?
 
#8 ·
Hi there

I'll try and figure out how to do it this evening and post results

Hth
 
#9 ·
Here you go, MY07 BMM 56k miles at 24.5 deg.C. These are what I logged at idle and increasing rpm. The boost pressures jump around a lot which I suspect is because the engine is free running and not under load? The highest peak value I saw was 1070mbar (15.5psi).

Rename the .txt to .csv and import into Excel.
 

Attachments

#10 ·
Hmm, that seems to be a different test. The one I do, the car revs up to 1400rpm, then alternates the boost valve between fully closed (duty cycle 10%) and fully open (duty cycle 90%) keeping the revs stable. It's in Basic Settings, not Output Tests - that may be the difference.
 
#11 ·
PM sent, that's an OBDeleven test, not found the same in vcds yet. :confused:

On a similar subject I occasionally get a boost pressure sensor fault latched. Well the fault code seems to be telling me the atmos pressure measured by the MAP sensor and boost pressure sensors aren't the same when the engine is stopped (sensors not tracking). Cleaning the boost pressure sensor with contact cleaner stopped the fault code for a few months, a new sensor made no difference and still the latched fault ocassionally comes back?

After some research others have found you can get a poor ground to the sensors. If the ground is bad causing millivolts drop along the loom ground wiring, the voltage (pressure) output will be raised making the pressure readings different. By how much before a fault code is set I don't know, but I think it must be small in millivolts. It is recommended to find the wiring loom ground connections under the air box, undo and clean them. I might also run additional ground wires in parallel to each sensor and splice them on to each connector.
 
#12 ·
I eventually found the boost pressure test which vcds calls charge pressure. When running the test my BMM engine maintains about 1450rpm and the boost pressure (N75) cycles up and down. For my test, the maximum boost was 1122.0 and the minimum 948.6 9 (which is near atmospheric pressure) The difference is 171.4 mBar which is a lot bigger than the minimum 80mBar low limit vcds suggests if those results are correct? I'll take it for a drive and log boost pressures with some hard wellie.

I reckon I must be well boosted then? :):)
 
#13 ·
Thanks - that tells me there's definitely something up with mine (either a boost leak or sticky valves, probably the latter), so it'll be off to my trusted local mechanic...
 
#16 ·
I thought you did all this yourself. :) I think I would put a real pressure gauge on the boost line to make sure the numbers I'm getting back from diagnostics are real?

I shall be posting some new info today on the boost pressure sensor which will be worth reading.
 
#14 ·
Hi there
I tried it last night. Figures were similar to Vox, but I ended test early as I didn't realise it would rev the engine as much. Though something was wrong!;)
 
#15 ·
The vcds test for 'charge pressure' was confusing to follow: According to the instructions you had to press the brake and accelerator together to start the test run. Nothing seemed to happen for my BMM engine.

Then I just clicked on 'enable test' and the glow plug warning lamp lit up (something was happening!). After a second the revs came up and stayed constant at 1450. Then I could see the boost pressure going up to a max and coming down slowly to the min.. I think it did this 2 or 3 times until it dropped to the starting atmos pressure. Then the engine revs dropped back to idle and the glow plug warning went out. I think the test lasted about a minute and you could easily get worried and stop it sooner. 1450 rpm shouldn't do any damage, but if yours was higher something is wrong?
 
#17 ·
Perhaps all this is due to a sensor G31 problem or its grounding? Is the sensor or wiring intermittent or putting out rubbish values and doesn't do it when stationary and diagnostics testing? These sensors are only checked at switch on for atmospheric pressure, who is to say the curve is wrong or there is something intermittent? In that case the numbers spat out by diagnostics are meaningless even though they may look 4 figure accurate to ten decimal places!

If you can power the N75 solenoid with a real pressure guage attached, that should tell you what the turbo can really do? If measured boost pressure readings don't correlate with diagnostics you have a sensor curve issue. If the maximum boost presssure cannot be reached you have a turbo\hose\valve problem. :confused:

I'm sure this isn't your problem, but people who get severe DIY tuning remaps done often forget about the boost pressure sensor. Its stock curve has only got so much overhead and will limit with the kind of over boost some tunes can do. :( In such cases, there are custom boost sensors available for a wider pressure range, but the ECU sensor table has to be recoded and would be done by a proper tuning shop. God help anybody getting their tune wiped at the stealers!

https://www.vweosclub.com/forums/showthread.php?t=35614
 
#19 ·
A mechanic you can trust who doesn't keep secrets to themselves and blag you with big bills is worth more than a 100 DIY attempts.

I always keep an open mind on electronically controlled closed loop engine systems, particularly if there's a lot of work and expense involved to remove and replace 'suspect' parts. I try to look for a second means of confirmation that doesn't need a computer. Unfortunately, the training most technicians get these days is to use and rely on diagnostics without understanding how things actually work. :(

As you discovered, the V.W turbo design is pretty clever and well explained in one of their training guides. It removes a lot of the turbo lag you used to get with earlier turbos because they can kick out the vanes at low rpm.

Let us know what problem was found.
 
#20 ·
Alright, it’s been a while, but I think I’ve fixed the issue.

An apparently common fault that causes the overboost fault to come up is the MAF sensor, and the test for this is really simple: just unplug it, drive and see if you get more power. The canonical test is to floor it going uphill in 3rd - you should get consistent power delivery through the rev range.

The test is simple - the sensor is at the top of the pipe coming out of the air box; unplugging the sensor will give you an engine fault but it’s safe to drive.

Replacing the sensor is simple enough -it comes built into the pipe it sits in, unless you have a five-point security screwdriver, so you have to remove the whole assembly. This just means unscrewing the air filter assembly, then unscrewing the MAF assembly from the air filter outlet, then detaching the MAF from the rubber air hose.

With a used part, the fix cost 5€, and the test cost nothing. Much better than a new turbo!
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top