While changing out my knock sensor yesterday, I noticed that only one wire (clear) attached to the knock sensor unit, yet if you follow the wiring harness up to it's all-weather connection near the driver's side front cylinder head, you'll see two wires (clear and black) going into the harness leading down to the knock sensor unit. The all-weather connector on the knock sensor unit doesn't seem to have provision for the black wire either. I removed the knock sensor harness and inspected it. I couldn't find the black wire anywhere except where it connects to the all-weather connector up top (by cylinder head) and goes into the sleeve of the harness. From what I gather, this black wire terminates somewhere in the wiring harness sleeve and is a "dummy" wire. I'm guessing since the knock sensor forms a metal to metal contact with the block, it ends up grounding itself without the need for the black wire.

Did anyone notice this or is there really suppose to be a black wire going to the knock sensor unit?

BTW, the all-weather connector (green) that you disconnect to gain access to the knock sensor "tunnel" IS NOT related to the knock sensor in any way. Look closely and you'll see that it's a MAP/BARO switch (note all the vacuum lines attached to it).

Dave

 

There is only one wire going to the knock sensor. I do not know why, but there is only one.

 

One wire goes to the KS, the black goes to ground. I believe it's on one of the bolts on the engine block up top near the intake manifold.

 

The black wire doesn't go to the knock sensor unit itself. It just disappears in the wiring harness and doesn't come out the end to connect to the knock sensor. I think it's a dummy wire because the all-weather connector for the knock sensor doesn't have a "port" for the black wire, only the clear wire.


Dave

 

Dave thats correct. From that weahter proof connector your looking at in the wiring harness, one lead goes to the sensor and the other goes to one of the ground bolts on the engine block. The ECU will see the difference in potential between engine ground and the small voltage signal the KS produces.

No signal is supplied to the KS from the ECU, the KS generates it's own signal. So from the ECU, it looks at real gound (engine block) and the sensor. Thats why only one lead goes to that sensor.

 

That's what I figured. Thanks.

Has anyone split open their knock sensor to see what exactly fails in them? There really can't be much to the part.

Dave

 

i have my old lying around...ill try opening it up, and see whats in to them...or if anyone wants, you can have it, just pay shipping/pick it up

 

The KS is a piezoelectric sensor - like a very sturdy microphone. Its has two terminals, the single wire and the metal base which provides a ground and closes the circuit.

 

exactly, it grounds itself.

 

Yeah it does ground itself, but that black wire is going to a ground on the manifold, wish I had a pic to show you, I found out by searching around the forum. Here is a site that explains it

http://www.motorvate.ca/auto/maxima/...ock_sensor.htm

 

The black wire is shielding for the clear wire.

 

You can try measuring the voltage response to vibration for the old vs new KS. I guess you can hook up a voltmeter to it while it is secured to something metal. Then probably drop or hit the metal to cause a vibration. As long as you hit it the same you can see how it affects both sensors.

 

DO NOT, I REPEAT, DO NOT DO THIS TO YOUR NEW SENSOR! YOU CAN DAMAGE IT BEYOND REPAIR!

Do it to the old one if you're curious, but please, this is a very bad idea to do this to a new one, unless you want to be in posession of TWO defective knock sensors.

The piezo-electric knock sensor is actually quite sensitive to vibrational shocks, especially harsh ones. It needs to be in order to perform it's job effectively. It's so sensitive, in fact, that the FSM recommends discarding one if it has been dropped on the floor.

If you whack a perfectly good K/S with a metal object, only bad things can happen. Hear me now, believe me later. . .


Use a digital multimeter and the testing procedures outlined in a FSM, Hayne's or Chilton's to test components, without destroying them in the process.

 

Actually I meant tapping the metal that the KS is bolted to . For example say one was to bolt the KS to an engine bracket, then you tap the bracket with your fingers or a screwdriver. It should produce voltage and then you can try it with the new KS. Sorry for the confusion.

 

I'd expect that you'd need an oscilloscope to see the voltage output of the knock sensor as it would be the same timescale as the shock, much too fast for the multimeter. I'd also bet that the output is pretty low - less than 1 volt. You can, however, see the principle at work cheaply - just buy a $2 disposable piezo-electric cigarette lighter. The crystal is fairly large and struck hard enough to get a spark to light the gas. Entertaining.