Direct injection dirty intake valves?

I was wondering if anyone is familiar with the apparent problem of very dirty valves on direct injection engines. I was made aware of this by a friend who has one of the newer turbo BMW 335i's. The problem is supposed to be caused by oil getting blown into the intake from the PCV system, and since there is no fuel ever cleaning the valves; this build up seems inevitable. This seems like a problem they should have run into when testing the engine... anyone have the lowdown on this?

Apparently this problem is happening to other DI engines. But here are the first images I saw..
Post for Carboned Valve photos - BMW 3-Series (E90 E92) Forum - E90Post.com

 
Last edited:
I would say most of that is due to excess crankcase pressure pushing oil down past the intake valve stem seals particularly when under boost, there was another thread titled 'PCV ( positive crankase ventilation saga...)' further down on the index page of that site where they had experimented with exhaust extraction valves mounted downstream from oxy sensor to help reduce crankcase pressure, also see that several are using 'meth' which I take to be methanol injection or possibly just water injection, that 'might' reduce the rate of carbon buildup, but remove the crankase pressure increase & that will remove the oil source & the problem in one hit.
 
Last edited:

Mike Trusty

GT40s Supporter
Lifetime Supporter
I know years ago BMW didn't use a PCV valve to restrict ventilation rate but rather just connected a large line directly from the intake manifold to the crankcase. This resulted in pulling high vacuum on the crankcase. Great for preventing oil leaks. Are you saying that is still the norm for BMW even with a turbo motor?

Even Renault knew better than that. They put the throttle on the inlet of the turbo on the R5 Turbo to insure a vacuum was always available for crankcase ventilation. THe only problem is that it over taxed the oil seal on the turbo shaft due to the high vacuum at off throttle. This resulted in massive smoke screens behind the cars when the seal failed. Ask me how I know.
 
The 335I has been an abortion since it's introduction. Lets see they have issued HPFP recalls SEVERAL times. They overheat when pushed. So much so that even the Intercooler is the first upgrade MOST people do (the BMW motorsport division actualy has this as a factory upgrade). The whole engine is crap and thus why most sell as soon as the warranty is up. This is the concern with the new M3 going turbo. BMW just has not gotten it down yet.
That said Audi is even worse. The whole S line is plagued with this. Search the forums and you will soon realize that this carbon buildup was a design flaw from launch and even Audi has tried to address this in several recalls and updates. Some helped but most didn't. Another reason people are dumping them once out of warranty.
 
Last edited:
It seems that many DI motors suffer from these problems, with some mfg claiming that drivers are using inferior gas, so they can't be held responsible for the issues...what a crock!

I believe Lexus have a patent on a system that uses direct injection as well port injection, which will clean the back side of the valve to some degree.

I love the upside of direct injection, but wonder if it's one of those "not ready for prime time" propositions.
 

Mike Trusty

GT40s Supporter
Lifetime Supporter
If this is a result of direct injection why does it not exist in diesel engines, turbo charged or not?
If they are pressurizing the crankcase with the turbo or not allowing it to relieve blowby then Jac has it figured out.
Of course the difference between a gas engine and a diesel is that the back side of the diesel intake valve NEVER sees a vacuum due to the lack of a throttle plate.
 
My mates V10 engined M5 is suffering from coked up valves etc (not sure if this is Direct injection mind??) and he does not spare the horses when it comes to driving the thing. Said he was in his local BM garage and they had the heads off an M6 de-coking them. Suspect it is to do with the Bio element in the fuel these days, I have noticed it a lot with the diesel we use in the tractors, EGR valves coke up for a pastime, sometimes in under 350 hrs.
 
There is an old saying, 'grass dont grow on busy streets'.... while most if not all of the cars in that article are likely to be driven quite hard, the reality is that they will also spend a large amount of time at low loads in traffic etc . Todays computers are configured to make cars run lean under these conditions which raises temps on parts like intake valves beyond what us older guys would call normal levels, hence the oil/crap buildup on the back of the valves. It would take a long time at full throttle to even start to remove that coke buildup, but remember in this case it would appear that those are the very conditions that allow the oil to coat the backside of the valve, then you slow down & drive quietly home & burn it onto the valve.

I remember an inlet guide setup for nascar that had the seal fitted at the port end, that might work, but probably not thru the whole warranty period.
 
I gather that some diesels are suffering from the same problem, I saw a pic of the head off some GM motor, intake port was completely stuffed with soot/coke etc
 
This reminds me of a wonderful comedy with Walter Matthau called 'A New Leaf ', wherein he drives a Ferrari 275 GTB around NY, continualy breaking down and finishing his journey behind a tow truck. ' Carbon on the valves' he says sagely to his companion. And we all thought those days had passed withthe coming of the ecu!
 
While its not hard to visualise why the oil doesnt get washed away since there is no fuel ( Solvent ) to do so, they [the owners] should be more concerned about the increase in crankcase pressure, I doubt BMW in this case use any tricks like keystone top rings or gas porting to increase the seal of the top rings, or whether all of this might be due to detonation in its early stages unseating the ring package, in which case its doomed to fail until they get a handle on that. Realisticly the mileages are reasonably high, dont see many real race engines acheiving 30000km plus without a major teardown!
 
The funny thing about the BMW N54 engine (the turbo one in question) is that its won all kinds of awards including the "International Engine of the Year" award. I'm starting to believe that these kinds of awards are a sham and these award-giving associations are just bought and paid for much like everything else. For instance, the VQ engine in my 350z has all kinds of awards and received much praise. But the same magazines that gave the praise just a few years ago are now saying its harsh and outdated...
 
I'm having trouble visualizing that--could you explain?
Nothing really magic about it, just the seal was fitted at the port end of the guide, the train of thought at the time I believe was that with multi stage dry sump pumps and crankcase pressures at speed that were 'below' atmospheric that the oil was not getting into the valve guide with the normal top seals & the odd engine failure was occurring from valve to piston contact after the valves seized in the guide.

To work in the case of the BMW with wet sump it would need a metered supply of oil to the guide & the 'top' of the guide isolated from crankcase pressure, in other words split the crankcase cavity pressure from that in the overhead gear with seperate breather tanks etc.

I suppose im really saying that if you can afford a BMW you can afford to pull it apart at regular intervals to de-coke it:)

Ive a 91 300zx, nice car & I like the shape, ...but... I really dont drive it much, darn cop magnet, but the electronic side of it drives me nuts...
 
Last edited:
I would say most of that is due to excess crankcase pressure pushing oil down past the intake valve stem seals particularly when under boost, there was another thread titled 'PCV ( positive crankase ventilation saga...)' further down on the index page of that site where they had experimented with exhaust extraction valves mounted downstream from oxy sensor to help reduce crankcase pressure, also see that several are using 'meth' which I take to be methanol injection or possibly just water injection, that 'might' reduce the rate of carbon buildup, but remove the crankase pressure increase & that will remove the oil source & the problem in one hit.[/QUOTE

Jac if the manifold pressure(under boost) is greater than crankcase pressure I dont see how it can pull oil past the VS seal.
I agree it looks like it is originating from the seal but maybe of boost (traffic driving with high manifold vacuum) .

Change one thing cause another ,I wonder if manifold vacuum is higher on DI engines as several manufacturers have the same problem.
Maybe the VVT is being rolled back for emissions and they are on zero overlap and causing high vacuum and the seals are not up to it.

Jim
 
Yes Jim, I tend to agree, it looks like BMW [ and probably others ] have a bit of work to do in covering all possibilities on this one. Most of those cars/photos in the original link also had aftermart modifications which probably cause conditions that the original computer might not be capable of dealing with though. Its almost like we are heading back to the early sixties when taking the cylinder head off for a de-coke @ around 30000miles was a regular occurance, difference is that task on a 4 or 6 cyl pushrod with two valves per cyl was relatively simple, not the case with this modern stuff where they are hidden under a wiring loom & plastic covers etc
 
Back
Top