The Effect of Intake Valve Carbon Buildup on AFRs in a 2.0L EcoBoost Ford

Hello everyone! I'd first like to start by introducing myself. I'm Tony Castillo, I'm from Chicago, IL and I currently own and am self-tuning my 2013 Ford Focus ST using the COBB Accessport and Accesstuner software. This is my first real dive into ECU tuning, so please bear with me as I'm bound to be ignorant of many aspects.

The Story:

I've been doing some mid-load spark curve tuning lately and during the review of my extensive datalogging I've noticed that cylinders 1 and 4 are more often than not just a tad bit less advanced than 2 and 3 before a knock event registers. This got me to thinking (as most things do). I've watched almost all of the webinars here and I'm particularly concerned in this instance with the individual cylinder fueling concept that was discussed in one of the webinars. After reviewing most every carbon buildup thread available in the Focus ST related forums, this is what I've come up with.

I think carbon buildup on cylinders 2 and 3 are leaning out cylinders 1 and 4.

For those of you that may not be familiar with the platform, this is a 2.0L Turbocharged, Direct-Injection engine.

It is widely accepted that the Ford OE model for airflow distribution to the cylinders is insanely accurate and the Focus ST community knows that. It's also well documented that, due to the central PCV location on the intake manifold, cylinders 2 and 3 are significantly more affected by carbon buildup than cylinders 1 and 4 are.

We also know that AFR is only measured by one exhaust source that cannot distinguish between individual cylinders as we are not lucky enough to have individual oxygen sensors for each cylinder's exhaust stream. Although this platform is blessed with factory wideband pre and post-catalytic O2 sensors.

I'm theorizing that as carbon builds on cylinders 2 and 3 inlet valves, the airflow becomes restricted and therefore causes those cylinders to get less than expected airflow, but they still get fueling for the expected airflow. They then fire what is now a rich mixture, it goes out the exhaust and the ECU sees the entire exhaust collectively as slightly rich. But, since it can't determine individual cylinder AFR, the ECU trims a small bit of fuel from each, evenly. This in turn causes 1 and 4 to now get a slightly lean mix while cylinders 2 and 3 get a slightly rich mix, but the end result of all the exhaust gasses combined is right on target, so the ECU thinks its hitting its target lambda.

Follow me?

I think this lean mix on 1 and 4 is causing those cylinders to be the ones (specifically 4) that end up having singular knock issues on WOT logs on this platform. Especially on tuned/modified vehicles running higher boost. They're just not getting the same level of fuel as they should be. I think that once carbon buildup becomes noticeable on cylinders 2 and 3, cylinders 1 and 4 are running lean, all the time, but the ECU has no way of knowing or learning this and has no way of compensating for it dynamically as carbon buildup increases.

The solution to this point has been to just pull the manifold and check the intake valve carbon buildup regularly (every 10k miles or so). If it's uneven, which it will be, it needs to be cleaned.

My real question here is... am I crazy? Is my thought process in line with what is really happening?

This is very likely the problem. The only thing you can do is physically clean ALL the intake valves using the Walnut-shell method. Then check your logs again. If its still knocking... then you get your injectors cleaned and balance tested. If its STILL knocking (unlikely) after cleaning the valves and injectors.... well then its just an inherent design flaw of the engine itself... and you'll need individual cylinder trim/correction.

The next time I go to change my oil, I plan on pulling the intake manifold and taking a look at the valves to see what is happening in there. The car now has 23,000 miles on it, I'm sure theres some level of build-up on the valves. It's very easy to get the manifold off of this engine, it only takes about 15 minutes. If they're covered, I'll clean them.

I should clarify that the car isn't knocking wildly. It idles and runs great. This is just an issue that we're starting to all see on the Ford EcoBoost engines now that they're getting some miles on them. I was more or less just concerned with the likelihood of my theory being correct. As far as individual cylinder trimming, I feel like that's almost a matter of chasing my tail because as time goes on and the valves continue to get build-up on them, won't the trim have to continually be adjusted to compensate? That seems like it's an unrealistic goal for the tune seeing as theres no way for it to dynamically adjust as mileage increases. Unfortunately, we might be stuck with having to clean the valves at some regular interval to keep this from becoming a much larger issue.

I think you misread my post. IF it turns out that its an inherent design flaw of the engine... THEN you'll need individual cylinder trim. But you'll only know that after you have cleaned valves and tested injectors. Having to clean the valves on a performance DI engine (especially one thats boosted) is just something you'll have to put up with. Its not a big deal at all... in my opinion. Especially not for an enthusiast who can do it all himself!. First thing i would focus on is implementing the best possible air/oil separation system.

You could always pull the manifold and give it a good clean and then fit a Good quality well designed catch can to seperate the oil from the air before being fed back to the intake to stop any buildup from occurring.

Marek, I see what you're saying now. There is a design flaw, but I'm not sure I would really call it a flaw. The factory manifold has the throttle body in the center and the flow path of air from the throttle to each cylinder seems very well balanced from the factory. The stock ECU also has cylinder compensation fuel tables in it calibrated for what I expect is a factory stock, completely clean engine. I'm assuming it is fairly accurate, as I can't believe Ford would go through the trouble of mapping it if they weren't going to do it accurately. The difference between cylinders isn't much at all according to those tables, and with a maximum deviation about about 2%. I think the 'flaw' is that the straightest path of least resistance for the oily vapor is a straight path to cylinder's 2 and 3. The PCV port, as you can see in the picture, is directly behind the throttle body and should distribute vapor equally to all cylinders, but it does not in the real world. I have an aftermarket manifold on my car, but the entire ST community is seeing the same distribution of oily vapor and buildup as I am.

I'm definitely not concerned with the effort of cleaning them regularly myself. It's not difficult and not very time consuming either. As far as the catch can idea, there are plenty of catch cans on the market, and none of them seem to be doing a good enough job of separating the oil from the air to keep it off the valves completely. Most of them just delay the build-up process. As far as I'm concerned, I'm not going to go through the trouble of spending the money and time mounting a catch can if I'm still going to have to clean these valves regularly. With the amount of miles I drive per year being around 25,000 I suspect this will be something I have to do every spring when the weather gets nice. I'm okay with that.

The overall thought process here is that this line of engines aren't just in performance, enthusiast marketed vehicles. They come in everything from mid-size cars to small SUV's and even the F-150 pickup truck. Every single one of these engines is prone to this in their own way. What about someone like my mother who's bought an EcoBoost Ford Fusion and is just going to drive it and never pay any attention at all? I'm expecting her's to give me loads of trouble once it gets up to the 60k+ mileage range if I don't go well out of my way to clean it.

Well if you're getting trace knock on those 2 cylinders then obviously the factory tune isn't that great... hence the reason why you started this thread!.

Its quite difficult to get the thermo & fluid dynamics equal throughout the cylinders. So technically, this is a "design flaw".

I'm sorry I must not have been clear. This isn't the factory tune. This is my tune on the vehicle that has been modified to run higher boost with aid of an aftermarket turbo-back exhaust and intercooler.

It did not knock while stock. And it didn't knock initially. I believe its only begun knocking now as carbon buildup has begun to develop.

I'm going to pull the manifold to investigate and the valve's condition.