Of course, you have less noise filtering if you turn it off, ASSUMING it was setup correctly in the first place. MS's documentation isn't very clear on this IMO, so I can't say if it is setup right. If turning it off fixes the problem, then for sure the nosie filtering is not setup correctly.

Had feeling this problem wasn't fixed.

I never use the noise filtering.

The larger gap didn't help. I think it actually made it worse. The engine is smooth and strong otherwise.

I left the larger gap alone (0.051") and turned off Noise Filtering and the problem went away.

Any reason to get in there and tighten up the gap?

Thanks much,

No, it's fixed. Enjoy.

Any speculation as to what might have happened? As I had mentioned, the tuner thought it was something he'd done, and loaded the previous tune. The problem remained, so we both thought hardware. The new Mazda sensor fixed the problem, but only by turning off Noise Filtering. The new CARQUEST sensor did not fix the problem, even with NF off.

Thanks for your time,

Like I said sensor too close to wheel probably caused the aftermarket sensor to not work at high RPMs. Mazda's sensor is more robust in this regard. Thus why i said test it while you had that sensor to verify if that was the issue.

Regarding noise filtering the noise filtering settings are wrong if enabling the filtering causes a loss of sync. Either figure out how to set them correctly (Good luck, I tried searching when you posted this thread and got nowhere) or just don't run it. I don't run noise filtering either as Scott says.

So, you think on the dyno, the installed sensor partially failed?