MOHFpro90

iTrader: (11)

I have had the 0304 code for a while now. So, I decided to test the KS and see if I could see what was wrong. Well, I didnt get any reading(just Overload) on my multimeter. I tried both pins of F121, and the GND of the battery. Any ideas? Does this mean my KS is completely shot, or did I just do something wrong? I was thinking I would get an awkward ohm reading if it was bad, not just nothing.

Sorry if this is a noob question, just this isnt my category.

vilmaxima

the code u r talkin about is P0304? If u used OBDII, thats a cylinder 4 misfire,
my ks sensor went bad also and noticed a heavy drop in acceleration and shifting(automatic). Maybe its a spark plug issue? Hope this helps

MOHFpro90

iTrader: (11)

No, the CEL code. Its my knock sensor, I know that. Im past that. Im trying to get some help on why when I test the sensor, I get a reading of 0 ohms. Anyone?

vilmaxima

knock sensor resistance is very high(use an ohmeter capable of measuring at least 10 M-ohms) check for continuity between 2 terminals of the knock sensor connector. Continuity should be indicated, if an open circuit is indicated, replace the knock sensor.
Got this from Haynes Repair Manual.

pmohr

iTrader: (26)

1, are you getting infinite or no resistance? i.e. 0 or 1?

What resistance range are you testing with, if not an auto-ranging DVOM? If you keep getting infinite, keep upping the test range. If you get no resistance, it's shorted somewhere, not likely internally so somewhere close. Of course, your meter could be shorted internally, too.

UncleMax98

Only one connector pin goes to the knock sensor. The other pin is a shield wire and is not connected to the knock sensor at the end. If you measure resistance between two connector pins, you will get infinite resistance. You can measure resistance between the pin and ground. One pin should be infinite resistance. The other pin should be around 525K Ohm. Note: Could be a bad cable, too.

SVI30

KS is a piezo-electric material, similar to ceramic. So you will not get anything on an ohm meter. The buzzer in a watch is made with the same material but function is just the reverse of a KS.

pmohr

iTrader: (26)

Hmm, and yet the holy grail itself tells us:



...I'd rather err on the side of Nissan, myself. Given that many many a person here have tested the KS with an ohmmeter and gotten an actual reading, I'd have to say that you're wrong. You will show resistance between the KS pin on the subharness and ground (assuming everything is as it's supposed to be).

EnervinE

iTrader: (11)

You're not using an ohmmeter that can read that continuity. The ohmmeter we had sitting around did the same thing; turns out it couldn't read that range.

OnEBadAsSi30

iTrader: (1)

A knock sensor doesn't set off the CEL its wire harness that sets it off becuase it recognizes that theyre is nothing there if the wire is corroded or cut.

BlackMaxdout

what? The ECU cycles through the different sensors every time the car starts up. If it finds one that isn't within operating spec, it throws a code. A bad KS itself will throw a code.

If you have a multi-meter that will read a resistance that high, and you get zero resistance, then yes, the KS is bad. Just go out and buy the $0.99 resistor that fools the ECU into thinking you have a good KS. Someone help me out here...is it 560K ohms? Just make sure you reset the ECU to clear the code, and ONLY run premium gas...even though gas prices aren't going down.

OnEBadAsSi30

iTrader: (1)

Okay then do this for me.....Go take the 10mm bolt out of your CKPS in front of your block. Dangle the sensor in front of the block. Try to start the car. IT wont start it will only crank but the car WONT throw a code becuase the car recognizes that the sensor is there but its not reading anything. A sensor can be out of spec and still work. I had a bad crank ref. sensor by the rules on the FSM but it still worked. The ECM automactically detects knock and backs off timing giving you horrible gas mileage and a huge loss in power. So go ahead and while the engine is running take an extension and hit the KS sensor. The car will automatically back off timing so the car wont knock. It won't throw a code. Now undo the KS harness and start the car. The car will start fine and run fine but it WILL throw a code because it's not there. My KS was "bad" even though it ohmed out outside of what the FSM said it should be. But my car gets 28 mpg and starts and runs just fine.

MOHFpro90

iTrader: (11)

Thanks, thats what I read in the stickies. I try to read them before asking a question. Not trying to be offensive at all man, but I come as prepared as possible. Thanks for the info though.

I went up to 20Mohms. I still got nothing(displayed OL, overload). Meter isnt shorted, it reads my gnd resistance fine.

I already mentioned neither pin gave me a reading. And again, I read the stickies.

Cable is good, thanks for the idea though.

Isnt 20Mohms enough? 20 million ohms? Or am I seeing something wrong?

Thanks guys.

BlackMaxdout

You're wrong again. It's a piezoelectric device, which means that the right torque is required to put the sensor within operating spec when bolted to the engine. When you unbolt it like you said, that torque spec disappears. And yes, it WILL throw a code, because it's not testing to see if the KS is "there", it's testing to see if it's getting the right resistance from it during startup. When the resistance changes from knocking, then ignition is retarded.

OnEBadAsSi30

iTrader: (1)

Okay then how come when I was doing my swap I threw a knock sensor code becuase I ripped apart the harness. But when I fixed it...It went away? But when I took out the sensor, and kept the harness plugged in I didn't throw a code. Same with the CKPS. I took it totally out. I threw a code for it. I plugged it back in and it went away. Can you try to explain that one?

NmexMAX

iTrader: (22)

Your car is special and does not comply with the other 4th gen rules. With that said, I have chocolate chip, vanilla, and peanut butter, take your choice.

arpangurjar

I am trying to test mine too, another method I have heard of is just disconnect and see if it makes a difference, if it dosent that means the sensor is bad, if it does the sensor is good.

Although I find this very crude method, might just work, at least it dosent harm the car for sure.

SVI30

The ECU is not looking for the resistance but the waveforms from the KS.

The piezoelectric material in the KS will generate a waveform when mounted on the engine and the engine is running. It transforms mechanical vibration into a voltage swing. So the ECU is looking for this waveform at startup. When knocking happens, the larger vibration translates into a larger waveform. The ECU input can detect the waveform voltage differences between a running engine and a running engine with pre-ignition occurring.

Conversely, if a waveform is applied to the piezoelectric material, it will transform this electrical energy to mechanical movement. That's how buzzers work, like the back-up warning beepers on trucks.

BlackMaxdout

Just out of curiosity, do you mean like a frequency waveform?

One of the only other sensors that I know of that creates a frequency waveform is GM MAFS. I know first hand, since I tried to use (and did, successfully) in my undergraduate project when I was in school. It was an air flowbench, and thanks to that MAF, it was damn near impossible to calibrate. I know, I know, this isn't 4th gen related.

Edit: For people who like to learn:
http://www.picotech.com/auto/articles/knock-sensor.html

NmexMAX

iTrader: (22)

MAF, frequency? No, MAF = voltage, not hertz. CPS = freq, VSS, RPM. We've been through this before, and I told you that the GM/s is not linear to V. Therefore it's not a good calibration standard.

Unless you share/plot the gm/s WRT RPM.

SVI30

Yes, 4G MAF outputs a voltage. That GM MAF uses a multi-vibrator and the frequency change is controlled by a resistor placed in the airflow. This resistor is on the film and heated. The increase in airflow reduce the resistor temperature and changes the resistance. Then the output frequency of the multi-vibrator changes.

Added:
The KS output is not a regular waveform. I do have some examples but no good place to post it. Since it is sensing cylinders firing, the frequency is tied to RPM. But the KS is not a perfect sensor, so the output looks like noise. Waveform of engine knocking looks like bigger noise.

lexoid

You are right and wrong at the same time Vibrations from the engine and engine knocking is transformed by KS (which is a piezoelectric element) into voltage. On the other hand, inside KS there's a resistance about 560 Ohms connected to the piezoelectric element. In this way ECU can check circuit to the KS telling that it is not open or shorted.

SVI30

That's interesting. Is this 560ohm resistor in series or in parallel to the piezoelectric element?

Added:
I just read the hyperlink in Post #19. That KS design has a 560 ohm resistor for diagnostic purposes. However, since it is in parallel to the KS, when measured with an ohm meter; meter will read 560 ohm since the KS itself is high impedance (almost like open).

The KS design in our 4G reads high impedance on a meter. So this KS doesn't have the parallel 560 ohm resistor inside.

I venture to guess the Nissan ECU looks for a waveform at startup to determine whether the KS is working instead of checking continuity via the parallel resistor.

For our engine, it is very convenient to have a oscilloscope on hand. For that, I have a cheap used scope at home originally bought to work on that GM MAF.

arpangurjar

I also am getting horrible milage, so I decided to test the KS by getting it out.

I wish I could post pictures but I guess i dont have rights, I am waiting, I removed my KS outside the car, and its cracked from more than 2 places.

But the thing is,I tried testing it using an ohm meter, and the 2nd pin reads 550 Ohms with it grounded to the engine block, and there is connectivinty on the 1st pin.

Now it looks like its been run over a car, but the resistances is being read, am I to assume that its ok ? I check the resistance between pin 1 and 2 and its also 550 ohms.

Can anyone check the resistance between pin 1 and 2 and tell me what it is?

I would really appreciate it.

BlackMaxdout

GM mafs DO output frequency....

We worked around the general motors MAF though. We ended up hooking it to a DSM MAF translater, then setting the offset to zero. Plus we had software that would do fast fourier transforms for us. Bingo.

BlackMaxdout

Didn't know about the multi-vibrator, but we were well aware of the heated element design. Our advisor was an engineer for GM back in the day.

Are you by chance a fellow engineer, mechanic, etc? You seem to know your stuff.

BlackMaxdout

Are you sure? There's a "mod" to get rid of the knock sensor completely by simply plugging a 560K ohm resistor into the female end of the connector...

I've done it, and it successfully prevented code 0304 from ever coming back, so I'm assuming it works.

SVI30

Yes, used piezo buzzers while designing watches years ago. Made an injector tester while working on that GM car. Now working on lasers and wireless semiconductors for money. But working on cars is still my joy, all seven of them, no Honda and BMWs. Many guys here still put me to shame.

SVI30

I found a web host for pix. Here are some material on what a KS output waveform looks like.




Note: ECM not only control timing but enrich the fuel mixture also to eliminate knock. So when knocking is present, gas mileage may suffer. Then running 87 octane may not pay. Best is to try a few tanks of different octane under same running conditions and see which one yields the best mileage.