You say that you went to the cat delete over a year ago and that there was no audible detonation as a result of the change. So, you can probably eliminate the test pipe as a direct cause.

I realize that it´ll seem like a baby step, but why don´t you swap the stock DV back in place of the aftermarket piece, just to eliminate that as another possible cause.

 

Most times unless its a seriously huge leak you wont even know you have one without doing a proper test. A boost gauge will tell you nothing and you will never hear them in the car during driving, typically.

But they can cause all kinds of drive ability issues that people will think its this or that and never do an actual proper leak test. Its the #1 test on my list when I am diagnosing an issue.....I like to get it out of the way because its cheap and easy to do.

You description of knock as metal sound certainly sounds right......I have heard it before in other cars and it sounds like marbles in a tin can rattling around. When you can hear it I would say its a pretty severe event.

 

You need to order a cable and log the car w/ evoscan.......if it is really knocking bad enough to hear you should see some pretty high counts in a log. That is the clear cut way to getting your answers for what/when/where questions to what is happening.....and the only way to prevent a catastrophic failure

 

Good thinking ...great minds think alike, because I did that this morning and drove the car a few miles, but I still get the sound. Now, maybe the ECU needs to "adapt" to the stock DV?...I dont know. I was planning on giving it at least a few days to see if there is a change.

 

You are correct about the cable...there is no other way, and to be honest, with cars like ours, you HAVE to have it in your toolbox. As far as checking for a boost leak, how do you do that when the car ONLY does it under load? (acceleration)...I suppose I answered my own question...LOG LOG LOG (unless there is another way...?)

 

So the "knock" event is only happening at the transition into boost under normal driving loads and part throttle?........

Also.....reset the ecu when you get a chance before you drive it next.

 

You have to pull the intake out, cap off the turbo inlet and use compressed air into the intake system to pressurize it. If you find a leak - fix it and repeat until you dont hear anymore audible leaks.....

There should be a good how-to somewhere out there to give you more details - Google

http://www.vfaq.com/mods/ICtester.html

http://www.dsmtuners.com/forums/arti...et-tester.html

 

Yes, and it continues until i let go of the gas (to deccelerate); as far as the drive load...it happens in all drive loads, as long as i am into boost "range". I will reset the ecu, maybe the deleting of the CEL's that I have been performing periodically, has screwed something up.

 

I am assuming that you have the stock boost control. But whatever. Check all of the boost hoses for tears, or leaks and assure that each hose end is secured to its respective nipple with a tight nylon ziptie.

 

Have you taken off or modified any of the exhaust system near the turbo recently? Such as removing and reinstalling, the exhaust manifold, O2 sensor housing(or the sensor itself), or the DP, or the turbo itself. Does your car have the factory Anti-Lag system? If so have you loosened the two banjo bolt fittings for the ALS in the exhaust manifold?

Could be as others have mentioned a boost leak, or even a pre-O2 sensor exhaust leak which lets unmetered air into the exhaust stream and the ECU tries to correct for the unmetered air. Could also be a leak in the boost hoses going to the BCS(or MBC). Maybe even something funky with the knock detector circuit.

 

This is all good advice...I will check hoses, bolts and see what I come up with. I am going to kick myself in the head if it ends up being something stupid like that!

BTW, I have not moddified or moved ANY exhaust system component besides the cat...nothing else has been touced.

 

A boost leak would cause you to run rich as well as a pre o2 sensor leak. Not likely a cause for detonation.

How long have you been in Greece and are you sure its not 100 RON? Sounds like gasoline.

 

Naw, A preturbo(or pre O2 sensor) exhaust leak is a weird number. Why? Let's say for example that an exhaust manifold gasket is blown out, or that there is a crack in the exhaust header. What is gonna happen?

Imagine that your car was tuned spot on with perfect AFR's, and etc. Then your car's header develops a crack on one of the flange welds. Since the gases in your header are pulsating from positive pressure to vacuum...then unmetered air gets sucked in through the cracked weld.

The O2 sensor will read this unmetered air as a lean condition. So, the ECU will enrichen the fuel mixture. Now since fuel has been added...the O2 sensor will see a too rich condition and the ECU will begin to pull out fuel like crazy.

All of this lean condition and additional fuel has nothing to do with the actual amount of air which was measured by your MAF sensor. So the fuel mixture will be leaned out again by the ECU. This will swing back and forth between too rich and too lean Ad Infinitum. When it swings too lean.... the car will knock. That's how it is.

 

That crack would still pull in outside air and be measured by the o2. Actually, the ecu would be running in open loop, not using the o2 sensor. (WOT)

 

I am talking closed loop/part throttle when the ECU compares the O2 values to the MAF values. The knock event would take place upon throttle tipin at about 3000 RPM on up. So, in the transition from partial to full throttle.