As we all know, GDI engines are prone to "valve coking." Every manufacturer is dealing with this issue. Some are ignoring the problem (GM) and others (Toyota) are working on ingenious way to fix the problem.

Yesterday, I removed the intake manifold on my Son's 2014 Ford Focus 2.0 and cleaned the backside of the intake valves. (Vary dirty, by the way, with 35k miles on it.). The carbon was really stuck on and it took a lot of soaking and scrubbing to get it off. A thought then came into my head: "If the valves were Teflon coated, would the carbon not stick to the valves in the first place?" Maybe, coating the valves with Teflon would eliminate or reduce the "coking" problem. I don't know. What do you think?

 

The operating temperature of the valves will exceed the useful range of Teflon and similar sorts of PTFE formulations but there are probably some ceramic formulations that might be somewhat useful.

GDI engines will quickly develop a cosmetically objectionable level of carbon buildup on the valve but the important question is how quickly this level impacts either the operating efficiency of the engine or its reliability. Some of the early European efforts, BMW in particular, had a major issue with buildup sufficient to significantly impact engine operation occurring at quite low mileage. But not every GDI engine builds up residue at a rate that will cause problems early on. The 3.6L GDI engine (one of GM's first) in my 2008 CTS had no noticeable drop off in power and the highway and overall fuel economy was the same when I traded it in with over 80K as it was when new. Its replacement in my 2014 ATS (same 3.6L engine) is showing no issues nor do you see a big problem with these showing up on the Cadillac forum boards.

I am not worried about the LT4 in my Z06, GM and other companies certainly took consideration of this potential problem as part of the design phase. A combination of port plus direct injection can help clean the valves but I expect in many cases it is used to help provide the smoother idle which is often a characteristic of port injected engines.

 

Throw a can of Techron in every 3000mi.

 

Wish it were that easy. A DI engine injects the gas into the cylinder. It never touches the back of the intake valve and therefore Techron would not help coking.

 

I agree. GM has put millions of DI engines on the road since 2007 and they are not noted for having a coking issue. If you do a Google search you quickly find the big complaint about some of the 3.6L engines used in the large GM CUVs (Acadia, Traverse, Enclave) is about timing chain failures. The last time I searched there was no mention of valve stem coking causing any problems.

Don't look for a solution that isn't needed.

Bill

 

I've had my intake manifold off numerous times to swap on various ported versions and have noticed nothing out of the ordinary on the backsides of my valves (2014 Z51 w/35k miles).

 

Great point Bill about not adding something that isn't needed AND can cause problems.

Much of that 3.6L timing chain failure can probably be attributed to the ridiculously long oil change intervals suggested when the 3.6 direct injection engine was introduced. If I had followed the OLM suggestions with my 2008 CTS it would have gone close to 20K between oil changes. Mine was changed much more frequently and I reset the OLM after every third change. After a couple of years there was a recall and the OLM received a new calibration with much more frequent change intervals and at the same time the timing chain and tensioners got an added warranty.

Mine never had an issue with the timing chain and I credit the more frequent oil changes from its initial in service date for avoiding that issue. A friend of mine showed me photos of the top end of his wife's BMW at 60K miles after she had followed the similarly long change recommendation from BMW and that was enough to convince me that synthetic oil isn't magic and excessively long change intervals lead to undesired buildup in the top end of the engine. Apparently BMW at that time did neither a good job with avoiding valve coking or avoiding lube related issues with the valvetrain itself.

I am on my second 3.6L (2014 ATS) and it is doing just as well as the first with a good mix of power and fuel economy. My LT4 engine doesn't have that many hours/miles yet but no complaints. And as soon as GM figures out the emissions CEL issue with the current Duramax diesel I will be ordering the replacement for my second Duramax equipped GMC Sierra pickup.

 

Although GM's GDI engines are not showing excessive coking deposits (unlike BMW and Audi,) their engines are still subject to coking. Here are pictures of the intake valve from my Z06 with 3,500 miles on it. At this point, I agree that the amount of coking is probably cosmetic and is having little or no effect on performance. Will there be an issue at 35k miles? There have been posts showing an increase in horsepower and torque after a valve cleaning, however (yes, those post are by vendors that sell catch cans.)

 

BMW friends do a walnut shell blasting every ~60k miles. Removes all the coking. Takes a few hours.

 

Direct injection cars should use top tier gas even my 2013 Buick Lacrosse owner's manual says so and it is also a DI car

 

That has no affect on this problem.

 

There are actually few engines that are non turbo that have issues with this with the one exception being the Lexus IS350. The LT engines really don't have an issue at all. I have walnut blasted hundreds of cars over the last 10 years and have found that turbo cars are far worse off than anything else.

 

Geez, are any of you old enough to have worked on engines before detergent oils were common? OMG! They often made these "dirty" engines look almost new by comparison. But, they often still ran pretty well up to being torn apart. I wish I'd kept some pix of the last 392 hemi (130k miles) I had apart that had never seen detergent oil. I think our new engines will handle all I've seen pictured here.