Alittle context to start, I have a new bottom end (IAG stage 2) I broke the 800 recommended mile range and had the car (2013 WRX 2.5l) tuned. After a few pulls my tuner grabs me and wants to shed some light on what is going on. His concerns were that the car was experiencing an abnormal amount of knock and was pulling timing, he didn't want to ignore too much of it incase there was an actual knock event. He pointed to somethings in the engine bay that should be corrected before doing the final tune.

My question is how sensitive is the knock sensor, would say a factory up pipe with broken welds on the heat shield cause enough vibration to have the output of the knock sensor trigger?

Another question about knock is have you ever seen fuel additives cause high levels of knock, and why is that?

Yes, any rattle will trigger that sensor. Anything not bloated and clamped down like stock is liable to rattle. Esp connectors and hoses and catch cans

If you add a bunch of octane and knock still these I’d suspect false knock.

I know the sensors are sensitive, i have an aluminum floor pan and it was tapping before we spaced it out and that was causing false knock and the car was pulling timing. so it's possible a broken heat shield could cause a false knock reading, idk if it is the case or not, but definitely possible.

so far as fuel additives, ive never run them so i'm not sure if they could cause knock. i woudnt think so but i just dont know. I have used torco for knock protection when racing, but that upps the octane rating of the fuel just to help prevent it, nothing changes on the tune when i run that.

Look inside for loose hoses touching the block, everything needs to be tight and away, make sure everything clamped, ziptied and secured.
Exhaust, make sure its has all hangers nice and tight, it can rattle bad and send vibration to the motor,
Just look everything over.
DO you have stock flywheel? Pulleys?
What RPM knock starting to show?

I have an exedy flywheel, its not lightweight I had CC lightweight flywheel and the chatter drove me insane. Stock pulleys.

Sounds as if I should run some high octane fuel to rule out false knock?..

I have my AP connected to my laptop right now in offline mode, I've got an older data log and a newer to make comparisons. Scrolling through both excel files, Columns L and M, feedback knock and fine knock learn are showing zero from top to bottom. Is that right, am I looking in the right place?

Get tank empty, fill up with best gas you can fine, drive around datalog, your AP should be able to datalog, as soon as you feel issue, stop log after that and restart, so you know where to find it later.

Your tuner sucks, find a real tuner. If all he is doing is relying on what a knock sensor is telling him and based on what he said "he didn't want to ignore too much of it (knock sensor) incase there was an actual knock event" tells me he doesn't actually know what knock is, and how to identify it during a tune.

There are so many ways to quickly determine this I would run from this tooner as fast as possible

Where the the knock occuring? What type of load and RPM?

I haven't been into boost with the car yet, so I haven't felt the car pull timing and fall on it's face.. I'm assuming I would feel the car buck and dip in power. Not sure I've ever experienced it honestly. As far as were the information is in the datalog what parameters should I be looking at??

I've also got new headers and an up-pipe on the way, I figured might as well get em both while I'm there.

I don't want to push the car until I know it's dialed. I broke the 1,000 mile mark today.

I think you should take it to someone else.

I've got the headers off the car at the moment, and when I took them down I noticed a thick dark liquid coming from cylinder one. I cant see very well but it looks like it all originates from the front most exhaust valve. I also noticed that the part of the casting has broken off from the other exhaust valve where the stem passes through. How f***** is the situation?

We need to see it, sounds like a valve guide issue?

I recently had a head job done when my new bottom end went in.. new valve guides were supposed to be installed where needed.. is that why the valve guides look tiered?