2013 WRX 2.5l misfire (not ignition) Pressure waveforms

Hi, I had a 2013 Subaru WRX 2.5l Turbo come in the shop last week with a misfire. The engine was replaced at 20,000miles with one that has 70,000miles. The engine was in a bunch of little pieces. Plugs were replaced a few months ago with NGK iridiums due to a misfire. Customer says the misfire would come and go. I did a current ram on the coils and they all looked the same. I did not have a paddle probe to look at the waveform. I scoped the injector and could see the pintle hump. Cylinder 1 under misfire data shows a roughness counter of 36. The rest stay around zero occasionally showing 1. Even though I usually don't swap parts I moved the ignition coil from cylinder 1 to 3 along with the spark plug and the roughness counter still showed cylinder 1 to be the problem. I also used a spark tester and the spark was able to jump a 30 thousands gap. I hooked a vacuum gauge up to the intake and it pulled 15inhg and and had little movement. It was not rock steady though. I put a pressure transducer in the cylinder and make some screenshots of the car running on cylinder 1 and 3. All 4 were taken when running but the ones with high compression were during cranking before the car started. I am wondering what the ripple is on the exhaust stroke. To me it looks like the exhaust cam may be advanced. I do not know how to know this for sure as I do not do in cylinder waveforms a lot.

I did a quick overlay on the running ones --- yes you may well be right on the cam timing but those ripples are really excessive.

I don't have another similar sized turbo flat 4 engine for comparison but this is a fairly typical known good 1.8 inline 4

I'm sorry guys, I didn't even realize when I did the overlay it didn't show the 4 lines. I tried to do it quick before I uploaded the photos. What would cause the ripples in the exhaust stroke? Valve guides? I can't find any info on it in any of my transducer books.

Nagging at the back of mind is that these engines are an oddball layout which could create anomalies in the manifold pressure pulses.

1-3-4-2 firing order so I would want to also look at in-cylinders from 4 and 2 and synch the ignition on to the incylinder.

I plan on doing that tomorrow. The customer insisted on driving the car over the weekend. It's just a complete pain to get to the other bank because the battery, air pump, and wire covers are in the way.

There was a Scooby thread on the forum a years or so back on a single cylinder low speed misfire ISTR it might have been on no1 cylinder, it was really odd but the cause was found to be a crankshaft trigger wheel issue --- strange that only affected one cylinder.

The waveforms labeled cylinder 1 and 2 are the passenger side bank. The ones labeled cylinder 3 and 4 are the drivers side.

Here are some better ones. I can upload the scope files too. I don't know if snapon changes the scales on the side. I tried to put the zero psi line at the large dash. Cylinder 1 seems low on compression. I wonder if the valve is moving on the seat.

I had a similar issue a while ago with a WRX. It would have weird codes and misfires that didn't make sense. In my case, it turned out to be a leaking fuel pressure regulator.

I only say this in that I don't always believe DTCs or misfire data. I try to go in cold while using the DTCs and misfire data as hints. LOL

Hey, I'm sure the scales are off, but is that really only 60psi of compression?

Have you done a leak down?

These are running compression waveforms not cranking.

So, two days ago I had a special guest stop by the shop. Brandon Steckler stopped by and was showing me some intake and exhaust pulse testing with my new pulse sensor. Michael Nicholson makes some very reasonably prices pulse sensors. Brandon was able to teach me how to diagnose the issue down to a leaking exhaust valve on cylinder one. That day was the first day I had the car where it sat cold. On a cold start the cylinder was completely dead. You can see on the waveform it does not appear to be an intake valve. On the exhaust pulse you can see the lack of a strong exhaust pulse. We confirmed this with a leak down test and the exhaust valve is definitely leaking. It took longer to do the leak down test then any other test. We also confirmed these tests can be done on a Subaru which it seems no one has done. This was questionable because of the split intake design. The brake boost hose can not be used since the check valve is built into the intake. Luckily there is a small centralized port we could use in the center front of the intake. Any questions feel free to ask.

Very cool. Thanks for the follow up!

DotterersAutoService wrote: Michael Nicholson makes some very reasonably prices pulse sensors.

As in, for sale? Or does he talk about building your own?

We confirmed this with a leak down test and the exhaust valve is definitely leaking. It took longer to do the leak down test then any other test.

Don't I know it. :silly: Just went through the same thing on a 2011 Suby with a leaking exhaust valve. Didn't feel confident enough in my own pulse waveform analysis skills to make the call.

Michael is on Facebook and makes them himself. I believe he makes them in batches of 30. A few of my friends and myself bought a set. I'll probably buy another to be able to do the exhaust and intake at the same time.