Infiniti G20 Nissan Primera Forum banner

1 - 20 of 181 Posts

·
Registered
Joined
·
390 Posts
Discussion Starter #1
Please tell me what in the world engine knock sensor is and what exactly it does?Just paid $300 to have it replaced(dealer could not clearly explain,but insisted it was very important and could possibly result in failed emission test)It better be important costing this money.Boy,did that hurt!
I tried to see if they in fact replaced anything or just reset the ECM( engine check light was on) but I have no clue as to where
the heck this sensor is.Please someone tell me I didn't waist $300.
 

·
Prince of Joy
Joined
·
4,614 Posts
The knock sensor is a small microphone attached to your engine that senses knocking, or detonation in the combustion chamber. When knock is detected, the ECU will go into 'limp mode' and your car will not be able to rev high and will run like crap. It is a failsafe that the engine enters to prevent you from blowing it up. If you had a CEL (check engine light) for a bad knock sensor, that is why you wouldnt pass emissions. You could have tried resetting the ECU just to pass, but the light would come back and if you ever got real knock (remote possibility) your car could go boom.


- Greg -
 

·
Registered
Joined
·
579 Posts
We never replace the knock sensors at my dealer. Because the check engine light will come on when knock is detected. Engine knock can be caused by alot of things like cheap, low octane fuel, carbon deposits on pistons and screwed up ignition timing. When the sensor detects knock it instantly sends input to the ECM and the timing is retarded to avoid any more knocking...it happens almost instantly and most of the time you cant notice it, but if you have constant knocking the sensor itself may be bad or the carbon deposits or ignition timing f'd up.
 

·
Registered
Joined
·
414 Posts
Hi,

I have a P10 G20, and I'm trying to determine the meaning of a code I am getting from the ECU, Code 34, Knock Sensor.
I tested the resistance on my knock sensor the other day, and I got .3 Ohms when I set the multireader at the lowest setting (200 Ohms). I think this is equal to 3 Ohms.

I don't know if this is good or bad or what number I should be seeing.
I did not see the specific resistance in the manual.
Do you know what it should be?

Does the engine need to be warm to test the resistance of the knock sensor?

Thank You
 

·
Registered
Joined
·
613 Posts
I fixed my k.s. code a couple of months ago. Theres tons of info on this, but I'll tell you the sensor itself is rarely the problem. The signal wire for the knock sensor is shielded(incased in a ground wire) and the signal wire shorts to the ground wire.
 

·
Registered
Joined
·
38 Posts
I fixed my k.s. code a couple of months ago. Theres tons of info on this, but I'll tell you the sensor itself is rarely the problem. The signal wire for the knock sensor is shielded(incased in a ground wire) and the signal wire shorts to the ground wire.
By any chance... Do you know where exactly that happens on the harness? I was going to route an alternative wire to get rid of the code but, I would love to give it a shot at fixing the intermittent short on the harness first.
 

·
Registered
Joined
·
613 Posts
It tends to happen where the injector harness splits and runs down the intake runner on high port motors. I'm not sure howLow port harness is routed, but the wire will be swelled up in the spot where the short is.
 

·
Registered
Joined
·
414 Posts
So, if the signal wire has shorted to the ground wire, will this effect acceleration at all, or is the 34 - Knock Sensor just a code that appears with no real effect on engine performance?
In other words, Will replacing this wire make my car run better? (It just doesn't feel quite as peppy as it should.)
 

·
Registered
Joined
·
613 Posts
^ when the computer picks up a ks code, it will retard ignition timing, so yes. It will affect the way the car accelerates. I was able to find my short, but there's nothing wrong with running a new wire if you can do it neatly.
 

·
Registered
Joined
·
414 Posts
I used an inexpensive multimeter to test the resistance from the KS terminal (a) to the negative battery terminal.
According to trbobrk (above), the reading should have been 500-620 Ohms. I got .3 Ohms when I had set the multimeter dial to the 200Ohms setting. .3 seems incredibly low and makes me wonder if the reading would have been difference had I removed the KS. I have not used a multimeter many times, so I am assuming .3 is .3 Ohms. What would make the resistance so low, assuming it is supposed to be 500-620Ohms? (The .3 was very consistent measurement. Accurate, I am not sure, but it was the same number every time.)
 

·
Registered
Joined
·
613 Posts
A direct short to ground will give low ohmage. If you unplug your K.S. you should have no continuity to ground (infinite resistance) on the k.s. circuit. If its unplugged and you still have continuity to ground, you have a short
 

·
Registered
Joined
·
414 Posts
You'll have to excuse my need for clarification and lack of knowledge.
So, when testing for resistance, if I touch EITHER of the terminals on the KS when it is still attached to the engine (but unplugged, obviously), the multi-meter should read Infinite resistance (1) without showing any number change. If there is a short the multi-meter will show some number, in my case .3, and this would indicate the KS has a short or is bad. Is this accurate?
By the way, Which setting withing the Ohms range should my multimeter be set to when testing the KS?
Thanks
 

·
'Luke'
Joined
·
5,319 Posts
We never replace the knock sensors at my dealer. Because the check engine light will come on when knock is detected. Engine knock can be caused by alot of things like cheap, low octane fuel, carbon deposits on pistons and screwed up ignition timing. When the sensor detects knock it instantly sends input to the ECM and the timing is retarded to avoid any more knocking...it happens almost instantly and most of the time you cant notice it, but if you have constant knocking the sensor itself may be bad or the carbon deposits or ignition timing f'd up.
Then what do you do ? I have a P0325 code but no CEL.
 

·
Registered
Joined
·
613 Posts
You'll have to excuse my need for clarification and lack of knowledge.
So, when testing for resistance, if I touch EITHER of the terminals on the KS when it is still attached to the engine (but unplugged, obviously), the multi-meter should read Infinite resistance (1) without showing any number change. If there is a short the multi-meter will show some number, in my case .3, and this would indicate the KS has a short or is bad. Is this accurate?
By the way, Which setting withing the Ohms range should my multimeter be set to when testing the KS?
Thanks
I said to check the k.s. circuit, not the knock sensor itself. I'll tell you one more time, THE KNOCK SENSOR IS USUALLY NOT THE PROBLEM. Unplug the knock sensor. Disconect the ecu. Find terminal 27 at ecu connector, (both ends of the circuit open) check for continuity between terminal 27 and ground. If you have continuity then there's a short in the circuit
 

·
Registered
Joined
·
414 Posts
Thank you for your help and clarification. I will test the circuit as described above and assume that the KS is good, but I do not know how to disconnect the ecu. That is what I am stuck on at this point. Could someone please tell me how to disconnect the ecu?
 

·
Registered
Joined
·
414 Posts
Thanks p10_swapper. There is continuity of 555K Ohms in the wiring to the Knock Sensor. I measured it from above at terminal 27 on the ECU connector, and I measured it from the disconnected knock sensor harness from below the engine. 554-555K Ohms at both places. Here is the view from below --> http://youtu.be/Ax5GrTl5Yaw
Power steering pump is to the right, auxiliary air regulator is just above the view of the camera.
You can see the knock sensor in the left of the picture (little, round, black thing with the rusty bolt in the middle of it).
If anyone knows what the brown thing to the right is, the thing that has a little yellow wire exposed, please let me know.
The electrical tape is loose at the knock sensor and at the brown thing. I had a leak in the power steering pressure hose, so most of the area and the wires got pretty wet from the p/s fluid.
I think I read a thread that suggests where the short happens most of the time.
I do not know how to install a new wire. I don't have a lift, but if it is something do-able, I would appreciate knowing what is required and how to go about it.
Thank you.
 

·
Registered
Joined
·
613 Posts
555kohms is infinite resistance. Answer something for me. were you testing with one probe on the k.s. wire and one probe to a good known ground such as a clean spot on the block or the battery? Because if your sure you did than it's not shorted. If you probed from one end of the ckt to the other and came up with 555k ohms, then there's a break in the wire. But for future reference 555k ohms means no continuity.
 

·
Registered
Joined
·
414 Posts
I tested the KS wiring twice. Both times I had the one probe on the negative clamp of a jumper cable running to the negative terminal of the car battery. One of the tests, I had the other probe touching terminal 27 on the ecu connector. The other test I had the other probe touching the interior of the female connector that plugs into the KS. On both tests I got 554-556K Ohms.
You are saying that this indicates the wire is not shorted.

To be sure, tomorrow I will do what you mentioned here, touch one probe to terminal 27 on the ecu connector and the other to the female connector of the KS. If I get the 555K Ohms again, I'll know there is a short in the wiring.

Thank you for clarifying that 555K ohms means no continuity. I was thinking the opposite was true. (that if the multimeter showed the 1 on the far left of the screen and did not change, this meant no continuity and if it showed any numbers this meant there was continuity. I now understand this to be incorrect.)

So, if a wire or device is grounded will it always have infinite resistance? And when there is no ground it will have continuity, or no resistance, (no reading fluctuation or change) when resistance is measured?
I appreciate the correction. Thanks
 
1 - 20 of 181 Posts
Top