Almost stock, 15 counts of knock

[Zenja]

I've datalogged the car to have 12-15 counts of knock at 10-12psi boost. The knock starts as soon as the car is spooled up and stays steady until I let off. It's the same at all RPMs basically. I'm trying to get some leads on what could be causing this to happen. It's been like this for as long as I've had the car and I need to fix it before bothering trying to mod the car faster.

Here's the information I know of. I do have fairly significant boost leaks that I'm trying to fix (I changed the TB shaft seals with the good, new kind but one still leaks a lot, I'll have to take it apart again). However, under boost, this would cause air to be LOST, so the car should run rich. In reality, the car seems to run lean, as there is no black smoke, and knock is usually caused by a lean mixture. This makes me think that something else is causing the knock.

All fuel trims are maxed out at 140%. This is probably (at least mostly) from unmetered air getting in through the leaks, but maybe not. Base timing is set properly. The car "lurches" at low speeds when going from no throttle to giving it throttle. What would cause the car to run lean? I wouldn't think it would be overrunning the fuel pump at this boost level and mods. I'll try rewiring it some time anyway though. I suspect the FPR might be the issue.

I've changed spark plugs, wires, and fuel filter recently. All sensors seem to be fine (looking at the logs) but the lower honeycomb on the MAS is removed. I'm getting fine airflow readings from the MAS, although I do find it odd that I can get them to max at 1600Hz with these mods. Injector pulse width gets fairly high sometimes but doesn't get near maxing out.

The car pulls fairly strong, but I know it should be faster. I ran it at 15.4@88mph here at ~1700ft altitude. Timing advance with this knock is only 9-10 degrees, so without knock the car should pull much better. Fixing the boost leaks will also make it better. Because of that I don't think there's anything dramatically wrong, but there's a small culprit somewhere.

Mods are as follows (basically stock):
- K&N filter, half cut off air can
- Test pipe
- Straight-through muffler
- Removed emissions stuff
- MBC, gauge

Thanks.

 

[sYnOnYx]

can you post the log please ??? that would help a bunch to see whats going on....

are you running premium pump gas ? 91 or better? ? ?

 

[slvrstingray]

Check to see if your "knock sensor" is melted.

 

[Zenja]

It's with 91/92 octane. There's no goo coming out of the knock sensor. I'll post a log later today because it's on my friend's laptop.

 

[Quasimondo]

The problem might be with your fuel pressure regulator. Try this experiment:

Take a pair of vise grips and use it to slightly pinch the return line on your fuel rail. This will artificially raise the fuel pressure. With the line slightly pinched, go on a run (tip: Zip-tie the vice grip to something very secure so it doesn't become loose at 60 mph). If the knock count is reduced or eliminated, then you have a leaking fuel pressure regulator that isn't raising the fuel pressure when boost is applied.

Also check the vacum lines between the intake, the FPR solenoid, and the FPR itself to make sure they're not leaking and preventing the FPR from raising the fuel pressure upon boost. If they look hardened/cracked/dry rotted, replace them.

 

[Zenja]

I saw you mention that suggestion when I searched. I'm going to try that tomorrow. The thing is though, I don't see how that would indicate the FPR is bad. Pinching the return line would raise fuel pressure anyway which means the car would get more fuel and richen out. There's probably a reason I'm not seeing though.

Here are 3 pics of different situations, all causing knock of varying degrees. I can put up numerical data of each of them if someone wants it.

You must be logged in to view this image or video.

You must be logged in to view this image or video.

You must be logged in to view this image or video.

I like this method because the pics are only 14kb and you can nicely see what's going on.

I should also mention, now that I look at my logs from about a year ago, the car used to knock a good ~5 counts less under the same circumstances as in the above logs. IPW and everything was the same, but the car now knocks more, and it's obviously running on the "low octane" timing table which is why my timing is so low. It seems like it's gotten worse over time, as it knocks more with less timing.