P0332, Knock Sensor Low Voltage

[chriskat01]

By the way, my understanding is that the resistance of the knock sensor to ground should be 100K ohms nominally. These are both measuring 98.7K ohms.

 

[rockola1971]

By the way, my understanding is that the resistance of the knock sensor to ground should be 100K ohms nominally. These are both measuring 98.7K ohms.

You should be measuring 93k-110k for a GM knock sensor. The voltage is a tricky situation. On most sensors you will have a power feed voltage and then a sensor output voltage and both of those voltages are DC. This is not true in the case of a knock sensor. A knock sensor only outputs a voltage when their is a knock condition going on in the engine. Here is the even trickier part that a DIY mechanic needs to know and remember when working on the knock sensor(s) circuit. The knock sensor is a piezoelectric accelerometer which is basically an old school microphone that uses a crystal that produces AC VOLTAGE when sound is acted upon it! So when you check the output of your knock sensor you have to either cause a knock (by tapping the block or head with a hammer) or the engine has to be running and has a knock condition (AKA Spark Knock). You wont see any DC output from the knock sensor. ONLY AC OUTPUT! During testing using the old hammer on the block test the knock sensor will produce around 1.3-1.5 VAC. Measure sensor out (wire that goes to PCM) using red lead of multimeter and black lead of multimeter on engine block (ground) or knock sensor body if it is out of the engine.

If both knock sensors test good for resistance (measuring at knock sensor harness) and both pass the sensor output voltage test then I would test wire(s) going to PCM. I believe it is a straight shot to the PCM connector from the connection to the knock sensor harness according to the schematic I attached earlier.

REMEMBER: Volatge coming out of knock sensor is AC VOLTAGE so put your meter on AC Volts! (NOT DC).

 

[chriskat01]

I thought I already posted this but apparently not. I understand that the resistance is supposed to be 100k ohms nominally. Both of these read 98.7K ohms.

 

[chriskat01]

You should be measuring 93k-110k for a GM knock sensor. The voltage is a tricky situation. On most sensors you will have a power feed voltage and then a sensor output voltage and both of those voltages are DC. This is not true in the case of a knock sensor. A knock sensor only outputs a voltage when their is a knock condition going on in the engine. Here is the even trickier part that a DIY mechanic needs to know and remember when working on the knock sensor(s) circuit. The knock sensor is a piezoelectric accelerometer which is basically an old school microphone that uses a crystal that produces AC VOLTAGE when sound is acted upon it! So when you check the output of your knock sensor you have to either cause a knock (by tapping the block or head with a hammer) or the engine has to be running and has a knock condition (AKA Spark Knock). You wont see any DC output from the knock sensor. ONLY AC OUTPUT! During testing using the old hammer on the block test the knock sensor will produce around 1.3-1.5 VAC. Measure sensor out (wire that goes to PCM) using red lead of multimeter and black lead of multimeter on engine block (ground) or knock sensor body if it is out of the engine.

If both knock sensors test good for resistance (measuring at knock sensor harness) and both pass the sensor output voltage test then I would test wire(s) going to PCM. I believe it is a straight shot to the PCM connector from the connection to the knock sensor harness according to the schematic I attached earlier.

REMEMBER: Volatge coming out of knock sensor is AC VOLTAGE so put your meter on AC Volts! (NOT DC)

Thanks very much, I knew a lot of that but for sure not the voltage that should be measured. I suppose in order to see this voltage I'll need to set my meter on peak hold?

One other question. I've already tested the wires best I can and they seem to be a dead short end to end and I can't make them ground to the chassis by moving the harness. What would everyone think of putting a new wire in place from the "main" connector up on the intake manifold down to the computer connector? I would solder both ends and maybe use an outdoor type shrink tubing to seal the joint.

Not sure what else it could be. Assuming the voltage from the knock sensors is actually good.

If it is a bad knock sensor this will be replacement number 3. That would make me not happy.

 

[Fless]

Unless you've already done this, I would recommend a very careful inspection of pin fitment, and cleaning (e.g., DeOxit or similar) of the pins on the ECM connector, before doing a wire bypass. This in addition to the voltage tests indicated earlier.

 

[rockola1971]

Thanks very much, I knew a lot of that but for sure not the voltage that should be measured. I suppose in order to see this voltage I'll need to set my meter on peak hold?

One other question. I've already tested the wires best I can and they seem to be a dead short end to end and I can't make them ground to the chassis by moving the harness. What would everyone think of putting a new wire in place from the "main" connector up on the intake manifold down to the computer connector? I would solder both ends and maybe use an outdoor type shrink tubing to seal the joint.

Not sure what else it could be. Assuming the voltage from the knock sensors is actually good.

If it is a bad knock sensor this will be replacement number 3. That would make me not happy.

No sense in fixing what aint broken. I suggest popping the PCM connector off the PCM and inspecting for corrosion. Green/Blue nasty things growing on the pin(s). During the hammer test on a knock sensor you should see an immediate jump of the AC voltage. You wont need a peak hold set on your meter. Just tap the hammer head on the valley cover plate and that should be enough to trigger the knock sensors. As Fless mentioned check for pushed pins (female and male) in the connectors too.

You can also depin the knock sensor harness and swap the 2 knock sensors wires around in the plug and see if the trouble code moves to the other bank. If it does then you have a bad knock sensor.

 

[RAMurphy]

You should be measuring 93k-110k for a GM knock sensor. The voltage is a tricky situation. On most sensors you will have a power feed voltage and then a sensor output voltage and both of those voltages are DC. This is not true in the case of a knock sensor. A knock sensor only outputs a voltage when their is a knock condition going on in the engine. Here is the even trickier part that a DIY mechanic needs to know and remember when working on the knock sensor(s) circuit. The knock sensor is a piezoelectric accelerometer which is basically an old school microphone that uses a crystal that produces AC VOLTAGE when sound is acted upon it! So when you check the output of your knock sensor you have to either cause a knock (by tapping the block or head with a hammer) or the engine has to be running and has a knock condition (AKA Spark Knock). You wont see any DC output from the knock sensor. ONLY AC OUTPUT! During testing using the old hammer on the block test the knock sensor will produce around 1.3-1.5 VAC. Measure sensor out (wire that goes to PCM) using red lead of multimeter and black lead of multimeter on engine block (ground) or knock sensor body if it is out of the engine.

If both knock sensors test good for resistance (measuring at knock sensor harness) and both pass the sensor output voltage test then I would test wire(s) going to PCM. I believe it is a straight shot to the PCM connector from the connection to the knock sensor harness according to the schematic I attached earlier.

REMEMBER: Volatge coming out of knock sensor is AC VOLTAGE so put your meter on AC Volts! (NOT DC).

Awesome post. I learned something today (therefore a good day). Always amazed with the talent and knowledge on this forum.

 

[chriskat01]

Sorry I haven't replied in a while, been working on some plumbing issues in the house. I've spray both connectors with cleaner today and repinned the connector to swap bank 1 and bank 2. We'll see what code gets set now, if any.

 

[chriskat01]

Okay, so set a new code on the ten mile trip to Lowes and back. Now it's P0327 which, based on the internet, is low voltage knock sensor bank 1. My reader doesn't give the bank for this code but it did for P0322. So I guess I have another bad knock sensor, this would be two bad out of the box and I happened to put the bad ones in the same spot. I'm thinking about buying two more attaching them somewhere they aren't intended to be but is easy to get to and hook them up, run for a few days to make sure I dont' get any codes before I change them, again.

Thoughts?

 

[rockola1971]

Okay, so set a new code on the ten mile trip to Lowes and back. Now it's P0327 which, based on the internet, is low voltage knock sensor bank 1. My reader doesn't give the bank for this code but it did for P0322. So I guess I have another bad knock sensor, this would be two bad out of the box and I happened to put the bad ones in the same spot. I'm thinking about buying two more attaching them somewhere they aren't intended to be but is easy to get to and hook them up, run for a few days to make sure I dont' get any codes before I change them, again.

Thoughts?

The problem moving to the other knock sensor wire shows it is indeed a bad knock sensor. Dont forget to swap your depinned wires back to their original place so you can keep track what a new code is pointing at later in life if another knock sensor were to fail. You can hookup a knock sensor but not install it. The only problem is that if you get engine knock then the sensors likely will not detect it and then your spark timing wont be retarded to stop the knock condition. Engine damage can happen then.
The sensors can be bench tested right out of the box on a workbench to verify a good part too.