Untraceable Ping/Knock/Detonation

This site may earn a commission from merchant affiliate
links, including eBay, Amazon, Skimlinks, and others.

Bad pinging sounds like ball bearings being rattled in a can, can't hear anything that sounds like it in the video
 
Hmmm.

The noise in the video (especially apparent towards the end with the on/off throttle at 3000rpm) does change volume and onset with timing - louder and earlier with advanced timing.

Just typing a stream of consciousness…

Intake side shouldn’t affect this. At one point I thought it could be a “gurgling” of sorts from the carburetor, but the noise has followed through 3 different carbs. An intake manifold crack is still feasible, but timing should still not affect any noise resulting from this.

We’re counting out things occurring inside the cylinder…. Unless the rings are vibrating / moving around with some kind of harmonic frequency from combustion? I would think it would run a lot worse and there would be a lot more oil smoke if this was the case though. The cylinder walls still have a very visible crosshatch from the rebuild too. I would think that any movement like that from the rings would mar the cylinder.

Like we said, I will adjust valves, but this has had no change in behavior in the past and timing still shouldn’t affect this.

As for exhaust- Timing could affect flow dynamics out of the exhaust and through the known leaks. I personally would think that later timing would make it worse with the flow being more volatile and possibly still combusting when entering the exhaust… but I don't have data or proof of this :)

Only other thing connected in this equation is the distributor. Possibly some mechanical noise or arcing somewhere in the distributor itself? But again, this is the second distributor it does this with.
 
Anybody in your area that's a gear head that could come listen? An old-timer? It's hard to capture noises like this accurately with a camera phone ... I've tried.
 
I've got a lifter tapping on a 5.3; 98% sure that's what it was anyway but was curious if chatgpt could do by listening to it. Here's what I got back:
Short answer: based on the audio, it sounds like a lifter/valvetrain tick, not a rod knock.


Why I’m saying that (from the analysis of your clip):


  • Transient (tap) rate ≈ 9.2 Hz at idle and it scales pretty evenly through the clip—typical of top-end ticks that track RPM rather than heavy, load-dependent bottom-end knocks.
  • The taps are high-frequency: median spectral centroid ≈ 5.6 kHz with a low-to-high energy ratio ≈ 0.12, meaning there’s much more energy in the higher bands. Rod knock usually shows a much stronger low-frequency thud.
  • Visuals: the spectrogram shows repeated bright vertical lines with lots of energy above ~2 kHz (a “clicky” signature) rather than thick low-frequency pulses.

I plotted your audio so you can see what I’m seeing:


  • Waveform and spectrogram are displayed above.
  • I also pulled a quick summary from the signal:
    • Duration: 9.0 s
    • Detected transients: 82
    • Avg transient rate: 9.22 Hz
    • Median spectral centroid: 5608 Hz
    • Median low:high energy ratio (≤800 Hz : ≥2 kHz): 0.12

What to do next (fast checks on a 5.3 LS):


  1. Warm vs cold: If it’s louder cold and softens warm, that supports lifter/rocker noise.
  2. RPM vs load: If it gets faster with RPM but doesn’t pound harder under load, that’s top-end. Rod knock typically deepens under light throttle/load.
  3. Stethoscope test: Use a mechanic’s stethoscope (or long screwdriver) on valve covers (top end) vs oil pan (bottom end). Your sound should localize at the top.
  4. Oil basics: Verify level and weight (usually 5W-30 on the Gen III/IV LS). A fresh oil change and quality filter can quiet a sticky lifter.
  5. Cylinder disable test: If you can (scan tool), briefly disable cylinders. Rod knock often changes/softens when the affected cylinder is cut; a lifter tick generally won’t.
  6. LS-specific note: If your 5.3 has AFM/DOD, a collapsed AFM lifter can tick like this. Persistent tick on one bank and occasional misfire at idle are clues.

If you want, tell me year/model (e.g., AFM or not), mileage, oil weight, and whether it’s louder on one side—I can tailor a tighter checklist (pushrod roll test, rocker clearance check, lifter soak, etc.).

Pretty impressive for a 10 sec clip from a phone recording imo.
 

Users who are viewing this thread

Back
Top Bottom