UDT999 High Knock Count and Fault Codes
#16
Possible causes for high knock
Faulty injector or injector lead (check for kinked cables, missing insulation, etc)
Faulty hall sensor will retard ignition and shouldn't affect knock (but will reduce power)
Faulty knock sensor will retard ignition and shouldn't affect knock (but will reduce power)
Faulty o2 will affect mixture and could lead to lean condition causing knock
Poor fuel quality
Poor tune (after market chip)
Faulty injector or injector lead (check for kinked cables, missing insulation, etc)
Faulty hall sensor will retard ignition and shouldn't affect knock (but will reduce power)
Faulty knock sensor will retard ignition and shouldn't affect knock (but will reduce power)
Faulty o2 will affect mixture and could lead to lean condition causing knock
Poor fuel quality
Poor tune (after market chip)
New Knock sensor today with no change whatsoever. Also drained 15 gallons of fuel out of the tank and replaced with 17.5 gallons of 93 octane fuel. At this time I added Chevron Techron fuel system cleaner and drove about 15-20 miles. Got home and checked the knock count and still 50%+ knocks and still getting rough vibration from engine at idle. I also noticed really rough vibration through car at low RPM 2000 - 2800 rpm. Definitely a misfire somewhere. With the injection valve faults I am guessing that these must be causing the "knocks" that are being registered by the software.
Also, O2 sensor is new, CHT sensor is new, 993 coils and modules are new, wires are new, caps and rotors are new, fuel filter is new.
Car ran fine before the dealership did the valve adjustment last November, with exception to noticeable low top end power.
Video here of engine running below. Do you guys hear knocks or pings? Maybe my ears are just bad. You can hear the exhaust blowing on the phone's microphone when I put it under the car that sounds like knocks, but that is simply "wind" noise. Car has a cat bypass and cup pipe....so exhaust is loud.
#17
Burning Brakes
Installed a Vitesse knock warning system. I got a few knocks at WOT around 4.5k RPM's on a SW chip. Put back in the stock chip and still had a knock driving normally. My engine is newly rebuilt with new knock sensors. Wonder if the system is just overly sensitive.
#18
Duck I considered this but 700 knocks out of 1200 ignitions is crazy. Must be related to the injector codes I am getting. Could this be a DME failure? Anyone have a spare DME hanging around that I could borrow?
#19
Three Wheelin'
I have had the shielded knock sensor wiring fail on a car before, i replaced the sensors and it still logged knock in the ecu. You can stub out the sensors with resistors of equivalent value, if it still shows knock on the ecu, you know its wiring or dme.
#20
Thank you for the suggestion. How do you go about doing this? Forgive my ignorance. I am a mechanical guy by nature, so electrical troubleshooting has me perplexed to say the least. I'm trying....so please be patient
#21
Three Wheelin'
The ecu will detect a fault if the knock sensors resistance is out of spec, but otherwise the ecu doesn't really know if the knock sensors are connected. This means you can test the dme/ecu or main harness wiring aren't the cause of the "knock" by replacing the knock sensors with resistors of an equivalent value - since you know it can't be seeing knock from the engine.
You'll need to test the sensors with a multimeter and find (roughly) the right value resistors to match. Then you can unplug the knock sensors in the engine bay, three pin AMP junior timer plugs iirc, one of the pins is the shield and you need to connect the resistor across the other two pins.
It's not elegant and you don't want to drive the car for any length of time like this but it does allow you to see if the "knock" signal is in fact just electrical noise due to failed shielding on the wiring loom - or, less likely imo, some internal issue in the dme.
You'll need to test the sensors with a multimeter and find (roughly) the right value resistors to match. Then you can unplug the knock sensors in the engine bay, three pin AMP junior timer plugs iirc, one of the pins is the shield and you need to connect the resistor across the other two pins.
It's not elegant and you don't want to drive the car for any length of time like this but it does allow you to see if the "knock" signal is in fact just electrical noise due to failed shielding on the wiring loom - or, less likely imo, some internal issue in the dme.
#22
The ecu will detect a fault if the knock sensors resistance is out of spec, but otherwise the ecu doesn't really know if the knock sensors are connected. This means you can test the dme/ecu or main harness wiring aren't the cause of the "knock" by replacing the knock sensors with resistors of an equivalent value - since you know it can't be seeing knock from the engine.
You'll need to test the sensors with a multimeter and find (roughly) the right value resistors to match. Then you can unplug the knock sensors in the engine bay, three pin AMP junior timer plugs iirc, one of the pins is the shield and you need to connect the resistor across the other two pins.
It's not elegant and you don't want to drive the car for any length of time like this but it does allow you to see if the "knock" signal is in fact just electrical noise due to failed shielding on the wiring loom - or, less likely imo, some internal issue in the dme.
You'll need to test the sensors with a multimeter and find (roughly) the right value resistors to match. Then you can unplug the knock sensors in the engine bay, three pin AMP junior timer plugs iirc, one of the pins is the shield and you need to connect the resistor across the other two pins.
It's not elegant and you don't want to drive the car for any length of time like this but it does allow you to see if the "knock" signal is in fact just electrical noise due to failed shielding on the wiring loom - or, less likely imo, some internal issue in the dme.
#23
Took the car to a local Porsche mechanic and he checked with his software (Durametric) and didn't read any knock counts. Must be something awry with my software or cable. Glad I had it checked before proceeding with additional troubleshooting and parts replacement.