Is direct injection a workable alternative?

[John Ranalletta]

While at the shop for an oil change, the tech related their shop is doing an increasing number of walnut blasting of head, valve stem chamber to remove carbon. He related that they've seen as much as a 50% occlusion of the ports in only 35-50k miles. Since there's no failure to operate, per se, dealers won't perform the service under warranty, but owners say they experience much better performance afterward.

 

The required, high-pressure fuel pumps have a spotty history as well.

 

 

Apparently, absent the cleaning effect of gasoline/air mixture passing over and around the intake valve allows carbon to build up; and, oil-laden recirculated exhaust gases add to the problem.

 

I have a similar problem with the 335d but worse as EG chokes the intake and head with carbon and oily goo. The tech inserted a wand cam into the intake and the picture is not pretty. Trying to decide whether to have the blasting/ultrasound cleaning done or trading it. The prospect of buying any car with direct injection is troubling.

 

I do lots of higher-speed driving (just finished 3k) round tripper to FL. Likely, if I didn't, it would have failed earlier as did a friend's 335d that was in the shop 2 months, awaiting a new head from Germany.

[MikeB60]

My daughter’s 06 GTI is a direct injection system and has about 95,000 miles. It’s had various misfire codes for the past 30,000 miles and there are numerous stories on various VW sites related to carbon buildup issues. I can get it to run better for a while with a good "Italian Tune-up" but that’s not fixing the issue. It's a fun car to drive but this issue has gotten to the point where it is more than annoying.

 

The high-pressure fuel pumps (HPFP) on the gas VWs have been somewhat reliable save the follower/lifter that drives the pumps of the camshaft. However the lifter is easy to check/replace and only about 20 bucks. The Diesel HPFPs on the VWs are another story.

 

Toyota puts an additional injector "upstream" to provide fuel to clean the intake system, don't know how effective this system has been in addressing the issue. BG has also developed a system to clean the intake track on the direct injection systems but I think this might help to maintain the system not to clear out a system with thousands of miles of built up carbon.

 

I understand the advantages of the system but i also think the carbon build up issue needs to be addressed in a method that doesn’t require 8 hours of work to disassemble the car every 30000 miles.

 

[John Ranalletta]

There may be a (literally) a solution in the way-back machine. It's called water injection.

 

I wonder why BMW thought it necessary on it's MotoGP safety car.

 

Water injection is the after-market solution to direct injection not working as designed.  Oldsmobile F85 had water injection in the 60’s.  Turbo Rocket Fluid.

 

One step forward, two back re: automotive technology. Think about it. To make diesels work, one has to put a water/urea mix in the tail and maybe, now, water and methanol in the snout.

 

 

[Guest Kakugo]

It's a very brand-dependent issue.

VAG engines are very prone to it, probably the most affected on the market. BMW and Ford engines are affected as well, but only some models. Sadly the big and powerful N57 diesel is one of them.

 

Proprietary Toyota and Hyundai engines seem, at the moment, the least affected: never heard a complaint about them regarding carbon buildup. Mercedes-Benz is also pretty good, chiefly on diesel engines: after all their prime lines of business are commercial vehicles.

 

There are some technologies in the pipeline that may just address this issue but (and here's a big but) it remains to be seen how much will they add to the tag price and much they'll detract from reliability.

 

 

[Guest Kakugo]

There may be a (literally) a solution in the way-back machine. It's called water injection.

 

I wonder why BMW thought it necessary on it's MotoGP safety car.

 

Water injection is the after-market solution to direct injection not working as designed.  Oldsmobile F85 had water injection in the 60’s.  Turbo Rocket Fluid.

 

One step forward, two back re: automotive technology. Think about it. To make diesels work, one has to put a water/urea mix in the tail and maybe, now, water and methanol in the snout.

 

 

Water injection?

 

B52G MITO

 

I am all for it, never mind the thunderous noise and the black smoke!

[Twisties]

Is direct injection different from standard fuel injection, or is this applicable to all fuel injected vehicles? Why are manufacturers going to this system?

[John Ranalletta]

In the illustration above, notice how the fuel is injection directly into the cylinder below the valves. In DI, the fuel is pulsed into the cylinder via a high pressure pump.

 

In traditional, port injection engines, like those on our boxers, the fuel is injected upstream of the intake valves and let into the cylinder when the valve opens; thus the fuel/air mix "washes" over the intake valve, keeping it clean of carbon buildup.

 

Mitch can better expound on likely benefits but from what I've read, DI engines are more fuel efficient for a number of reasons, but the jury's out on the carbon build up. Car makers don't test drive power plants over 3-5 years of cold starts, driving short distances. Just because an engine can run 500k miles on test stand doesn't mean it will last long with Aunt Tillie's driving habits.

 

 

 

[Twisties]

Thanks, John.

[smiller]

Most diesel engines are direct injection but they don't usually have any inherent problems due to carbon buildup (or at least not more so than you'd expect for any internal combustion engine.) Not sure why the apparent discrepancy between gas and diesel regarding this problem.

 

[Joe Frickin' Friday]

Is direct injection different from standard fuel injection, or is this applicable to all fuel injected vehicles? Why are manufacturers going to this system?

 

Very early fuel injected vehicles (e.g. 1984 Corvette) used throttle-body injection, i.e. 1 or two injectors dumping fuel in at the throttle plate. Basically a carburetor with computer-controlled fuel metering.

 

Before long everybody went to port fuel injection: each cylinder had its own fuel injector spraying fuel more or less at the backside of its intake valve. Fuel injection pressure is typically about 50 psi for these systems. Vaporization, atomization and mixing are dependent on heat coming from the port walls and intake valve head, followed by the violent inrush of air during the intake event.

 

For gasoline direction-injection (GDI), the injector delivers fuel directly into the combustion chamber, just like a diesel engine. The injection timing varies depending on what is being asked of the engine; injecting late in the compression stroke requires much higher injection pressures to finely atomize the fuel and also to disperse it across the chamber, thus the need for a high-pressure fuel pump that can deliver somewhere around 3000 psi. This is more than port injection pressures, but quite a bit less than diesel injection pressures (~30,000 psi).

 

As John noted, unlike a port-injected engine, the intake port in a GDI engine does not receive a regular bathing by gasoline, so any particulates in the incoming air may accumulate there. This includes not only dirt that gets past the air filter (no filter is perfect), but also soot that arrives with the EGR gas. GDI engines tend to produce more exaust particulate matter (PM) than port-injected engines, so it's kind of a double-whammy: the EGR includes more soot, AND the intake tract isn't getting cleaned by a fuel bath.

 

Soot formed during low-temperature combustion tends to have more organic-carbon content, as well as raw fuel and byproducts from incomplete combustion. Contrast this with high temperature combustion, which produces soot with more elemental carbon content, i.e. less sticky/gunky. Also, airborne soot particles are subject to thermophoresis, a process by which they tend to migrate across the warm flow and settle on cold surfaces. All of which makes me wonder if the severely occluded intakes that John mentioned in his OP are perhaps associated with engines that do a lot of short-trip driving?

[John Ranalletta]

Probably did lump two separate issues (GDI and crankcase oil). Some more adventurous owners are installing catch cans and water injection to cope with each/both. This is a gas engine.

 

 

This is a picture from my diesel intake.

 

 

In either case, walnut blasting the heads and ultrasonic cleaning of the intake is in order $700-1200 depending on garage.

[mneblett]

GDI engines tend to produce more exaust particulate matter (PM) than port-injected engines, so it's kind of a double-whammy: the EGR includes more soot, AND the intake tract isn't getting cleaned by a fuel bath.

 

Soot formed during low-temperature combustion tends to have more organic-carbon content, as well as raw fuel and byproducts from incomplete combustion. Contrast this with high temperature combustion, which produces soot with more elemental carbon content, i.e. less sticky/gunky.

I also wonder whether there is a crankcase ventilation aspect to this -- a bit of extra oil vapor drawn into the intake can result in a bunch of oil deposits. Corvette owners with LS-1 and later motors often install oil separating cans in the breather circuit to minimize the oil in the intake (which can get to the point of actually pooling in the intake). Not so sure about the EGR aspect -- diesels also use EGR, and combustion isn't always complete on every stroke despite the higher combustion temps (as shown by the power-chip guys that belch black fog banks when they stand on the throttle). Interesting discussion; subscribing.

[John Ranalletta]

Mark, that combination is exactly what's troubling the BMW diesels and you can see it in the picture of my intake manifold. The EGR valves soot up as well. This is a 335d egr valve (not mine) w/ 100k miles:

 

 

When these systems get to a certain point, the engine goes into limp mode. With 88k miles on mine, with the car still under warranty, do I continue as if the problem didn't exist and hope it fails under warranty, risking it dies while I'm on a long trip? Or, do I spend the nearly $1k for a cleaning?

 

BMW is no help. They know there's a problem and have developed a cleaning routine instead of replacing the heads. The car referred to in the quote belongs to a client. It carboned up, went into limp mode, had to be towed and was in repair for months waiting for a new head. The owner negotiated a dealer, not BMWNA, extension of his warranty.

 

Good news on the "time of repair". Since that car has been here ( mid-January), BMW has developed and approved a new repair that is vastly cheaper and quicker. No comparison. His will be one of the very last in the US to have the old repair. The part that we were all waiting on will not be needed for the new repair. Yay!!

 

but are willing to do nothing in advance of the car dying from the problem.

 

Your BMW ESC Platinum contract DOES cover this issue on your car. As long as the car is symptomatic. IF you were getting ready to go out of that extended warranty, time or miles, I'd suggest you contact us/me and see what/if your options are, assuming that your car is not (at that time) yet exhibiting symptoms. This is not a maintenance contract, and will not cover items that might "break", but only items that ARE an issue on YOUR car.

 

I understand the extended ramifications, which is why I'm saying to contact us before you go out of extended warranty.

Love the car, but think it's my last BMW.

[roadscholar]

It's a very brand-dependent issue.

 

Mercedes-Benz is also pretty good, chiefly on diesel engines: after all their prime lines of business are commercial vehicles.

 

 

This is what I'd assumed knowing BMW gasoline engines have had the problem for some time (not aware the diesels did). So I just asked my local MB tech (factory trained and works at a large Mercedes dealer) if their late model diesels were also getting carbon buildup in the head. His answer, "Yes, and in the EGR system. Only ones I see that stay relatively clean are ones using ultra low sulphur fuel only and normal oil change intervals".

 

This next isn't terribly relevant but I found it interesting the origin of mechanical fuel injection came with it's own set of problems.

 

From Wikipedia:

 

First direct injection[edit]

 

1955 Mercedes-Benz 300SL Gullwing Coupé

 

Mercedes-Benz 300SL Roadster

 

1956 300 SL

 

1956 "Gull wing" open

Like the W194, the 300SL borrowed its 3.0 litre overhead cam straight-6 from the regular four-door 300 (W189 "Adenauer") luxury tourer introduced in 1951. Featuring an innovative diagonal aluminum head that allowed for larger intake and exhaust valves, it was canted to the right at forty-five-degrees to fit under the SL's considerably lower hoodline.

In place of the W194's triple two-barrel Solex carburators, a groundbreaking Bosch mechanical direct fuel injection was installed, boosting power almost 25% over the Gran Prix car's. Derived from the DB 601 V12 used on the high-powered Messerschmitt Bf 109E fighter of World War II, it raised output from 175 hp (130 kW) to 215 hp (160 kW), almost double that of the original Type 300 sedan's 115 hp (86 kW).

The result was a top speed of up to 260 km/h (161 mph) depending on gear ratio and drag, making the 300SL the fastest production car of its time. However, unlike today's electrically powered fuel injection systems, the 300 SL's mechanical fuel pump would continue to inject gasoline into the engine during the interval between shutting off the ignition and the engine's coming to a stop; this unburned gasoline washed lubricating oil from the cylinder walls, which not only left them unprotected in affected areas during start-up but would dilute the engine's entire oil supply if the car was not driven hard or long enough to reach a sufficient temperature to evaporate the gas out of the oil.[citation needed]

Exacerbating the problem was the engine's large racing-oriented oil cooler and enormous 10 liter oil capacity, which virtually guaranteed the oil would not get hot enough. In practice, many owners would block off airflow through the oil cooler and stick rigidly to the appropriately low 1,000 miles (1,600 km) recommended oil change interval. Clutch operation was initially very heavy, remedied by an improved clutch arm helper spring which reduced pedal force. From March 1963 to the end of production later that year, a light alloy crankcase was used on a total of 209 vehicles.[3]

Aerodynamics played an important role in the car's speed, with Mercedes-Benz engineers placing horizontal "eyebrows" over the wheel openings to reduce drag. Unlike many cars of the 1950s, steering was relatively precise and the four-wheel independent suspension allowed for a reasonably comfortable ride and markedly better overall handling. However, the rear swing axle, jointed only at the differential, not at the wheels themselves, could be treacherous at high speeds or on imperfect roads due to extreme changes in camber. The enormous fuel tank capacity also caused a considerable difference in handling depending on the quantity of fuel on board.

 

 

[smiller]

I also wonder whether there is a crankcase ventilation aspect to this -- a bit of extra oil vapor drawn into the intake can result in a bunch of oil deposits. Corvette owners with LS-1 and later motors often install oil separating cans in the breather circuit to minimize the oil in the intake (which can get to the point of actually pooling in the intake). Not so sure about the EGR aspect -- diesels also use EGR, and combustion isn't always complete on every stroke despite the higher combustion temps (as shown by the power-chip guys that belch black fog banks when they stand on the throttle).

The problem (with modern emission-controlled diesels, and perhaps gasoline engines as well) is not so much EGR or oil in the intake singly, but rather the combination of the two. A properly-engineered EGR system alone shouldn't be too problematical, and engines have been burning a bit of oil for 100 years without too much drama, but add soot and oil together and it's a recipe for a mess. There are many popular mods out there (especially for diesel models) to add catch tanks or vent PCV systems to the atmosphere instead of the intake manifold, and also delete EGR in the ECM programming. This stuff used to be done for mileage/performance (on which it had little if any effect) but more and more it's being done to avoid intake contamination problems as the engine ages.