Intermittent knock measured when returning to idle - potential cause?

Hi everyone, i am at my wits end as i chase knock when i feturn tonidle from a WOT pull. Not always by once or twice a time. Any possible reason for this? I always thoight KNOCK happens at hard running of the engine.

My ride is a Miata NA6 ITB engine on ME442 - AlphaN tune.

Any advise?

Thanks in Advance

does it knock at wot ? or when allready clsing throttle , closed throttle would seem very unlikely to me due to the lack in combustion pressure

have you listened actually or are you reading ecu data ?

a/ are you SURE it's "knock" from the engine and not something external to it?

b/ if it's detected by your electronics' sensor, is it possible you have it set to a too-sensitive setting?

c/ what, if any, method are you using for load, or "vacuum" advance? Is it possible it's over-advancing under the higher vacuum on over-run on closed throttle, especially from high rpm?

d/ is it possible it's running excessively lean for a "lean knock"?

As an aside, if you could please edit your post, so it isn't so painful to read?

Lots of good questions asked.

I'm curious if you're detecting knock as soon as you tip out or right as idle RPM is achieved.

As Kenny asked, I'm curious if you're listening to it or reading data.

To throw a few ideas out there for now with the little bit of info we have:

ITBs make all sorts of goofy noises near idle.

I don't know if you're running an engine with VVT but they can rattle or clang during sudden drops in oil pressure or sudden large swings to a phasing stop as the engine drops to idle.

Sorry about the gibberish as I was hammering it away on a iPhone.

I am detecting knock by the ME442 attached knock sensor and such that the ECU kicks in Knock evasive actions and initiate timing reduction as set up in the ECU.

I am getting knock indication when I lift off my foot of the pedal, meaning as it goes from 80% TPS to 0% TPS.

With everyone's input, I am thinking that the registered 'knock' could be the throttle plates slamming shut and making a 'ping' for the knock sensor to supposedly think it is a knock event. I will try to adjust the throttle plates to not shut fully to test this hypothesis.

One other knock event I am seeing on the data-log is during starting, there are days I get no knock indications other than during startup. Any possibility of knock event on cranking?

I am planning/contemplating on booking a dyno session to go through the Ignition Timing vs TPS vs RPM table cell by cell, using TunerNerd to listen for knock and pushing timing for MBTT. I was thinking if I have no knock events during dyno, then any other knock event picked up just by the ECU-data logging should be false knock, correct assumption?

Thanks for response from everyone.

Looking at the MEITE software it looks like the knock setting could do with some extra settings to prevent the throttle closure 'knock events'.

It might be worth bringing this up with the guys from ME, they're usually quite quick at implementing any easy changes to their firmwares

It sounds like there's room for improvement in knock detection as I would not expect detonation during startup.

Knock not occurring on the dyno can occur otherwise due to changes in atmospheric conditions, fuel quality, vehicle issues popping up, different loading, perhaps heat soak etc. so I wouldn't ignore everything that happens off the dyno as false knock, but having that data from the dyno certainly helps you make your best judgement.