Using Waveforms to Identify a misfiring cylinder.

I had a fun one come in, this week. I thought I would share it.

1999 Ford F150, 5.4L.

Noticeable, somewhat intermittent misfire. Misfire Monitor flags no misfires. No Codes. Fuel trim data indicates the misfire is on Bank 1.
Spark Plugs, Coils, Injectors, Fuel Filter, and Air Filter have all been replaced. Relative Compression looks, relatively, good. How would you tackle it? Here are a couple, different ways.

This waveform contains all the information needed to identify which cylinder is misfiring.



RED = Crank Signal
GREEN = #1 Primary Voltage


If we add Crankshaft Frequency, (BLACK Trace) by creating a "Math Channel", we can see the Crankshaft slow down, due to the misfire.
*ignore the HUGE drop-out in the BLACK trace. That is caused by the missing "Sync Notch". The Math Channel interprets the missing tooth as a BIG speed reduction.





Adding a Piston Position Chart, we can see the Frequency of the Crankshaft increase with each Power Stroke, except Cylinder #2. The Crankshaft slows down during #2 Power Stroke, indicating insufficient combustion.







This image contains all the information needed to identify which cylinder is misfiring AND indicates which Valve is at Fault.


BLUE = Pressure Transducer in the Tailpipe
GREEN = Cylinder #1 Ignition Coil Primary Voltage
*This is a RUNNING @ Idle Waveform*


Using a Piston Position Chart, we can see that when the #2 Exhaust Valve opens, the BLUE trace falls into a vacuum. This indicates a misfire on Cylinder #2. (Remember the Old-School "Dollar Bill Trick"?)





The BLUE Trace has another anomaly, too. Notice that, when Cylinder #2 comes up on Compression, Tailpipe pressure, slightly, rises instead of the normal oscillation. This indicates that Cylinder #2 exhaust valve is leaking.








A Picture is Worth a Thousand Words.

Here is a link to the .PSDATA file, if anyone is interested.

drive.google.com/file/d/1g-_p5PPNMu2u5E9...YF2/view?usp=sharing

Impressive, as always Chad!

I'm having a hard time seeing the frequency change in the black trace, it looks to me as if it lines up with the sync notch every time.
Would you care to elaborate on that aspect of the waveform a little further for me?

I'm wishing more and more that my Snap On scope could keep up with that Pico software...

Noah wrote: Impressive, as always Chad!

I'm having a hard time seeing the frequency change in the black trace, it looks to me as if it lines up with the sync notch every time.
Would you care to elaborate on that aspect of the waveform a little further for me?

I'm wishing more and more that my Snap On scope could keep up with that Pico software...

Like Chad said ignore the sync notch dropouts themselves. The frequencies either side of them shows the rpm dropping on the bad cylinder.

Noah wrote: I'm having a hard time seeing the frequency change in the black trace, it looks to me as if it lines up with the sync notch every time.
Would you care to elaborate on that aspect of the waveform a little further for me?

This is a picture of the Frequency Math Channel at it's default Range (zoom level).
The Black trace is the Crank Frequency. The upward spikes (RED) can be ignored. With some filtering on the Crank signal, these will disappear. The YELLOW circles are the Sync notches. They are, merely, bumps at this zoomed out range.
The over-all Frequency of the Crankshaft signal is 389.1 Hz (GREEN).




If we edit the Frequency Range to a Min/Max of just below/above the overall frequency, we can Zoom in. The upward spikes (RED) can still be ignored. The HUGE downward spikes (YELLOW) are the Sync Notches and should be ignored. Comparing the little (yellow) bumps, from the above image, to the HUGE (yellow) downward spikes in this image, show how far in we are zoomed. Ignore the sync notches.




This next image is zoomed in a little farther on the frequency. It has some filtering on the crank signal. This eliminated the upward spikes. Here, we can clearly see a repeating upward pattern (GREEN), indicating Crank acceleration. In RED, we see the downward Deceleration, due to the misfire.






Zooming in farther and adding the Piston Position chart, we can see that the pattern repeats its downward deceleration at Cylinder #2 Power Stroke (RED) . All other Cylinders show an increase (GREEN) in frequency during Power Stroke.

What I have circled in PURPLE looks like the frequency dropped from Cyl #8 to Cyl #1. It didn't. This is an illusion created by the sync notch. Ignore the sync notch.

Noah wrote: I'm wishing more and more that my Snap On scope could keep up with that Pico software...

Save your money. Buy it, even if you can't afford it. You won't regret it.

Still, love my Verus, too.

Thank you for the explanation. I couldn't unsee the synch notch.

To me it looked like the only reason there was an appreciable drop in the CKP frequency was from the sync notch and the pattern just ratcheted up until the sync notch appeared again.
Especially since the the sync notch and subsequent drop out aligned with the suspect #2 cylinder.
The zoomed out and annotated capture helped me see it much more clearly. I still don't know that I'd be able to pick that out of a live capture to make a positive diagnosis.

I've said it before and I'll say it again, you are the man Chad!

Thanks, Noah! But, I am not the man.:side: I am learning, like everyone else. In fact, I had the same problem that you did with the sync notch. A man named Paul (not Danner) turned me in the right direction. He posted this link:

Especially since the the sync notch and subsequent drop out aligned with the suspect #2 cylinder.

This messed me up, too. I, originally, thought that I had a misfire on Clinder #8, too. Paul (not Danner) helped me see my error.


I had only heard about creating a frequency trace with a Math Channel to identify misfires. This was the first time that I have used it.

I think it opens up a new door for MAF sensors, too. I'm itching for a MAF problem.:lol:

I still don't know that I'd be able to pick that out of a live capture to make a positive diagnosis.

No one could. You need to save the capture and, then, analyze it using a Lay-over, Cursors, and Zooming.

Awesome Thread Chad !!! lol

This is a 2016 GMC Sierra 5.3 @ WOT climbing a 7% Grade.


MAF maxed out @ 7.5kHz which is spec.

"Paul"

Weycraze wrote: Awesome Thread Chad !!! lol

"Paul"

Thank you, Paul (not Danner)!

And, thanks for your help!

I remember that CTI class - It was AMAZING. :woohoo:

I'd happily watch more from Adam Robertson. He's actually got a class in the upcoming ATE Expo:

www.atetrainingexpo.com/training

LMAO, Tyler! I love your Avatar! :lol: :lol: :lol:

Chad wrote: LMAO, Tyler! I love your Avatar! :lol: :lol: :lol:

HE JUST LOOKS SO ANGRY. :lol: I can't wait till Paul swings by and notices.

Tyler wrote:

Chad wrote: LMAO, Tyler! I love your Avatar! :lol: :lol: :lol:

HE JUST LOOKS SO ANGRY. :lol: I can't wait till Paul swings by and notices.

That green tint to it looks like he's turning into the Hulk :lol: :lol: :lol:

Matt T wrote: That green tint to it looks like he's turning into the Hulk :lol: :lol: :lol:

I was thinking Shrek. :lol:

I'm really wanting to sell my verus, it has the 18 software and it is a bulky tablet. I think I prefer my surface with the pico scope but I'm really holding on and will try to use my 2000 version of the pico for now. My job will end this month and I want do to automotive full time . The reason I want the Pico is because it can be trusted as opposed to other cheap scopes which sometimes you are not sure about and have to use another gage just for kicks.