haltech knock signal log @ 100hz looks like a staircase - is this a sign of something broken or incorrectly configured?

Heya - i have an e2500, Haltech knock sensor, ground shielded wiring (wired as per Andre's advice about where shielding goes etc) and I'm logging the knock signal using the internal datalogger @100hz.

But it looks more like a staircase than something like a wideband o2 signal would....

Have I stuffed up setting up the knock threshold table axis or something like that - or does it look like a faulty knock sensor?

They are costly enough that I don't want to just buy one to test things (strict budget at the moment dang it) - but if it looks to be a borked sensor I'll quite happily spend the coin on a new one.

I cannot believe that the noise profile on the engine would result in a log like this for an analogue sensor.

I imagine that flat line is the average or peak noise in each sample period, to actually see waveforms log sample rate would need to be high kHz range, you aren't going to see ringing just have an indication of sound energy which will be good enough to have some insight, probably best to induce some knock at low rpm & load to see what energy it delivers.

Haltech quote (backing up Slides comments above)

After checking with our firmware engineers, what you are seeing in the knock signal data is correct.

The ECU will ‘hold’ the last peak recorded knock signal value for a period of 500ms in order to make logging and analysis of this channel easier. At this time, the raw knock signal value isn’t brought through to the front-end for logging, however I have requested that this be added in a future firmware for better analysis of the knock signal.