Knock acceptable while "cruising", how much?

there is no knock?

Gosh I thought for sure there would have been more than 4 positive knocks, but a good portion of that time was steady speeds. I see knock most often when shifting through the gears. I realize this is small activity but sometimes it does seem more prevalent that what this shows.

Is there anything in here that would explain the knock, or does it look pretty random? I'm a novic at reading these logs and have tried to find patterns but I cant seem to. I did zoom in on a portion that shows some knock activity here. I do notice that the car goes rich when rolling back into the throttle after a shift. Is this normal stuff?

https://datazap.me/u/manbearpig/extended-lo-drive-work?log=0&data=1-8-14&zoom=376-1458

That's 4 events that it was positive for more than a fraction of a second. Zoomed out looking at 30 min total data zap doesnt have the resolution to show everything on that scale. Its pretty random as far as I can tell.

markmurfie said:

That's 4 events that it was positive for more than a fraction of a second. Zoomed out looking at 30 min total data zap doesnt have the resolution to show everything on that scale. Its pretty random as far as I can tell.

Click to expand...

Does it not seem to be related to the car going rich as I roll into the throttle after a shift? I'm curious if what I'm seeing here is normal? If this log looks pretty good otherwise would it be a bad idea to have the knock sensitivity turned down just s fraction?

I had to download your .csv and put it in to SC livelink to make a couple histograms.

Doesn't seem to be related to AFR. just low RPM and load.

markmurfie said:

I had to download your .csv and put it in to SC livelink to make a couple histograms.

Doesn't seem to be related to AFR. just low RPM and load.

Click to expand...

That looks like an alarming amount of red when you break it down like that, lol. Thanks for the insight. I know you're much more well versed at reading logs than I am. Do you see anything out of whack here? Other than the obvious small knock events? Do fuel trims and AFR pictures look good?

Nothing looks obviously out of whack. The amount of small events you have is strange, but they are very small.

The only correlation I can find is low rpm and low load and over 95% of the positive events are when the IMRC are closed or just about closed. I cant look at the slope value like in HPT, so I cant determine in which direction the IMRC are going when the events happen. My guess is when they open or close the knock sensors are picking the noise up as a very small knock event. Unfortunately HPT doesnt have much advanced functionality when looking at .CSV files not created by their scanner.

markmurfie said:

Nothing looks obviously out of whack. The amount of small events you have is strange, but they are very small.

The only correlation I can find is low rpm and low load and over 95% of the positive events are when the IMRC are closed or just about closed. I cant look at the slope value like in HPT, so I cant determine in which direction the IMRC are going when the events happen. My guess is when they open or close the knock sensors are picking the noise up as a very small knock event.

Click to expand...

Could I feasibly unplug both IMRC solenoid just to see if the knock goes away? Or would that do any good? I don't have a great understanding of how those things work.

What I don't understand is the inconsistency.

What was the road surface like? Seeing as the knock occurred at low speed events, possibly near junctions and low speed turns etc, were there pot holes or speed bumps?

I noticed that I see very short knock events on on my daily back road trip to work on swo specific c corners where the road surface is particularly bad. Is your car running a stiffer than stock set up?

If I remember correctly when I unplugged mine at idle they opened. It didn't really seem to effect much at idle. I did not try to drive the car like that so I cant say how it would drive. You should be able to drive with them unplugged and IMRC open. It may not drive exactly like stock, but the tune is setup to deal with situations where they fail open or fail close. As long as their position sensors are still plugged in.

It's definitely not road surfaces. I'll see the knock rolling into throttle after a shift on a smooth on-ramp.

I'll try to disconnect the cmcv and see what happens. That's the two blue electrical connectors at the back of the manifold right?

If your CMCV solenoids are bad, that could explain some of the other odd things you've posted about in this section of the forum, as well.

Man I hope so. That'd be great but I don't think I'll be so lucky. It still seems too coincidental to me that the knock is almost always right at throttle tip-in and the car also goes rich here. Seems like it would be related but I really don't know if that's normal or not. Mark, I think you mentioned something about an open loop to closed loop transient period. Is that what this is when the car goes lean at fuel cutoff under decel then immediately goes rich when rolling back onto the throttle? Pretty sure the stock tune does this as well if I can recall but maybe there's something mechanical going on if that's not a normal reaction.

ManBearPig said:

Man I hope so. That'd be great but I don't think I'll be so lucky. It still seems too coincidental to me that the knock is almost always right at throttle tip-in and the car also goes rich here. Seems like it would be related but I really don't know if that's normal or not. Mark, I think you mentioned something about an open loop to closed loop transient period. Is that what this is when the car goes lean at fuel cutoff under decel then immediately goes rich when rolling back onto the throttle? Pretty sure the stock tune does this as well if I can recall but maybe there's something mechanical going on if that's not a normal reaction.

Click to expand...

Yes any rapid change in airflow whether when opening or closing the throttle(transient) needs to be accounted for in an open loop. Waiting for the mixture to get to the sensors and give you closed loop feed back would take to long and fuel corrections wouldnt be applied quick enough. Now if your CMCV plates are reacting slowly cause they are sludged up or the solenoids are not operating as quick as they should, then yes this open loop transient could be adding too much fuel as the expected increase in airflow is not actually as high as it should be.

I tried comparing your AFR with IMRC voltage and load, also with RPM. We know its rich on tipin so when looking at min values its going to mostly show tipin. When looking at max values it will show 2 as thats what it goes to in DFCO. It doesnt seem to go to DFCO when they are open, and it should above 3000 RPMS on decel. Theres also an interesting rich line at 2750 but at no other RPMs.

Thanks for putting in the work to put this stuff into tables but this is getting way over my head very quickly. Is it not normal to go rich on tipin like I'm seeing? I don't think it's a tune issue as the stock tune reacts this way too.

If there is something else off I'm going to have a tough time finding it. Can't exactly take it to a dealer for a warranty claim and say "hey I know the car drives great but the fueling is off during transient periods, could you replace the cmcv or anything else that could cause this".

I'll unplug the cmcv connectors and see what happens but I'm not holding my breath for any real improvement. I'm open to suggestions on what to try, though.