Knock Sensor ???

pcb

The Diagram Dude
:
2002 MP5
I Tested My Knock Sensor !!!

I was curious about my knock sensor and didn't like the look of the "goo" dripping from it.

027_zpsdbdc63cb.jpg




After reading about the knock sensor and it's parameters, I was concerned that my knock sensor may not be sending data and that my ECU may not recognize it.


knocksensor_zps18e1a0e3.jpg





So I decided to try to "fool" my ECU by replacing the knock sensor with a 560 kohm resistor to see what would happen.


051_zps2721ca11.jpg


057_zpsed0ffb45.jpg





I went for a drive and while watching my UltraGauge for engine load, temperature and the five second rule, my CEL did go off (code 0325),... proving that my ECU does indeed monitor for actual data.
I didn't notice any "pinging" that I would expect even during the first 4 seconds before the ECU reduced the timing advance because of the code. I'm guessing that may be because my engine is in great shape with no carbon buildup (regular seafoam treatments), no oil burning and I passed emissions with almost NO measurable pollutants (1 or 2% of what is allowable)

I then pulled out my resistor and got the same result.



HOWEVER,... I am still concerned about the goo dripping from my knock sensor. I'm thinking that the "goo" is the suspension that the accelerometer floats in (if it's not an "air suspended" crystal) and if that is the case then the sensor may be giving reduced signals due to the lack of free movement within the sensor.

Our knock sensor only has one CEL code,... newer vehicles have many more.

P0325 Knock Sensor 1 Circuit Malfunction (Bank I or Single Sensor)
P0326 Knock Sensor 1 Circuit Range/Performance (Bank 1 or Single Sensor)
P0327 Knock Sensor 1 Circuit Low Input (Bank I or Single Sensor)
P0328 Knock Sensor 1 Circuit High Input (Bank I or Single Sensor)
P0329 Knock Sensor 1 Circuit Intermittent (Bank 1 or Single Sensor)
P0330 Knock Sensor 2 Circuit Malfunction (Bank 2)
P0331 Knock Sensor 2 Circuit Range/Performance (Bank 2)
P0332 Knock Sensor 2 Circuit Low Input (Bank 2)
P0333 Knock Sensor 2 Circuit High Input (Bank 2)
P0334 Knock Sensor 2 Circuit Intermittent (Bank 2)


http://www.youtube.com/watch?v=0xI94ZDF54Y



My advice is that if you are experiencing any pinging,... definitely consider a new knock sensor. An old "gooey" sensor may be giving just enough signal to not set off the CEL but not enough to stop the pinging.


They aren't too much money so it might be worth "throwing money at" to find out if it stops your pinging.

knocksensor22_zps724e7c54.jpg
 
Last edited:
Pcb,
Would you consider the beck/arnley model? Or the other two higher priced ones?
 
On good gas, normal tune you should not 'need' a knock sensor. But good to replace just in case ;)

I am not sure a resistor would give the multiple 'counts' the ECU would expect to 'see' and then adjust timing. On my ECU (Lancer) a knock count below 4 does nothing. A count above 9 is cause for concern...

A single resistor signal may be just 'one count' to the ECU.
 
I am not sure a resistor would give the multiple 'counts' the ECU would expect to 'see' and then adjust timing. On my ECU (Lancer) a knock count below 4 does nothing. A count above 9 is cause for concern...

A single resistor signal may be just 'one count' to the ECU.

My interest in the knock sensor started when I read this thread:

http://www.mazdas247.com/forum/showthread.php?123826915-Need-Help-identifying-a-sensor

I was worried that a count of zero may be seen by the ECU as OK. I thought maybe as long as there is about 56,000 ohms of resistance across the sensor then the ECU wouldn't pop a code even if there was no data at all.

I guess I proved that the ECU does indeed check for some sort of data. (assuming my resistor was making proper contact).

I think my curiosity will take me to another step and unscrew the sensor and wrap it up in pink insulation and stuff it in the engine bay somewhere. That should effectively prevent any sort of signal from getting to the ECU and hopefully I will pop a code. If not, then I would assume that no data flow from the sensor is accepted as normal by our ECU.

If I don't get a code when my knock sensor is effectively muffled then I would worry about an old sensor sending weak data and not getting the ECU to back off the timing in a knock situation.


Interestingly, I couldn't get my car to "knock" even during the first five seconds before the ECU called my bluff and backed off the timing advance,... which makes me think that my sensor was sending no data regardless.
 
Sorry to bump an old thread, but when I replaced my knock sensor with the Beck Arnley one I noticed the one there was the same mitsubishi part I got from Beck Arnley. After looking at a picture PCB posted of his knock sensor, it also looks like he has the same mitsubishi part. And the pinging I was having has gone away.

Just an FYI for those that want to replace theirs.
 
I was curious about my knock sensor and didn't like the look of the "goo" dripping from it.

027_zpsdbdc63cb.jpg




After reading about the knock sensor and it's parameters, I was concerned that my knock sensor may not be sending data and that my ECU may not recognize it.


knocksensor_zps18e1a0e3.jpg





So I decided to try to "fool" my ECU by replacing the knock sensor with a 560 kohm resistor to see what would happen.


051_zps2721ca11.jpg


057_zpsed0ffb45.jpg





I went for a drive and while watching my UltraGauge for engine load, temperature and the five second rule, my CEL did go off (code 0325),... proving that my ECU does indeed monitor for actual data.
I didn't notice any "pinging" that I would expect even during the first 4 seconds before the ECU reduced the timing advance because of the code. I'm guessing that may be because my engine is in great shape with no carbon buildup (regular seafoam treatments), no oil burning and I passed emissions with almost NO measurable pollutants (1 or 2% of what is allowable)

I then pulled out my resistor and got the same result.



HOWEVER,... I am still concerned about the goo dripping from my knock sensor. I'm thinking that the "goo" is the suspension that the accelerometer floats in (if it's not an "air suspended" crystal) and if that is the case then the sensor may be giving reduced signals due to the lack of free movement within the sensor.

Our knock sensor only has one CEL code,... newer vehicles have many more.

P0325 Knock Sensor 1 Circuit Malfunction (Bank I or Single Sensor)
P0326 Knock Sensor 1 Circuit Range/Performance (Bank 1 or Single Sensor)
P0327 Knock Sensor 1 Circuit Low Input (Bank I or Single Sensor)
P0328 Knock Sensor 1 Circuit High Input (Bank I or Single Sensor)
P0329 Knock Sensor 1 Circuit Intermittent (Bank 1 or Single Sensor)
P0330 Knock Sensor 2 Circuit Malfunction (Bank 2)
P0331 Knock Sensor 2 Circuit Range/Performance (Bank 2)
P0332 Knock Sensor 2 Circuit Low Input (Bank 2)
P0333 Knock Sensor 2 Circuit High Input (Bank 2)
P0334 Knock Sensor 2 Circuit Intermittent (Bank 2)


http://www.youtube.com/watch?v=0xI94ZDF54Y



My advice is that if you are experiencing any pinging,... definitely consider a new knock sensor. An old "gooey" sensor may be giving just enough signal to not set off the CEL but not enough to stop the pinging.


They aren't too much money so it might be worth "throwing money at" to find out if it stops your pinging.

knocksensor22_zps724e7c54.jpg

To PCB thank you so much because this thread post really helped me out to Pin point the Location of where the Knock sensor wire is located. No body on this Forum was able to answer it the closest answer i got was yeah its up by the driver side Lol! This picture shows it all and yes i did buy the Beck Arnley one.
 
This has been on my bucket list for some time - I do get occasional mild pinging so I'm curious to see if it goes away with the new knock sensor. I too thought I had an oil leak last year and discovered the "black goo" thread. The stuff stained the back of the engine and around the oil filter. I was due for an oil change today so I jacked up the car, drained the oil, removed the filter and also found it helpful to remove the bracket for the dynamic chamber too. A 15/16" wrench works great - same size needed to hold camshafts... don't ask how I know that... LOL. Timed my engine so many times I bought the proper wrench size. Found it a bit fussy to feed the wire up in the correct place so as to avoid contact with dynamic chamber (hot). It will be interesting to see if there is any change in performance between that and the new front O2 sensor I put in to replace the original. The replaced (original) O2 sensor ate a lot of oil last year and Torque readings were intermittent. Suspect it was partially pooched.

Did a brain reset with the battery to start fresh.

Both parts came from Mazda - the O2 sensor was expensive, even with my trade discount. Ouch. Both the removed and the replaced knock sensor were branded Mitsubishi.





 
Last edited:
LOL I never could get the wrenches to stay right - plus one of my crescents is too big for the job and the adjuster knob kept gouging the head. I did the job so many times I invested in some camshaft holders as well - they're fiddly too because of the shape of the side of the engine under the spockets. It was a combination tie-wrapping the 15/16 to the strut bar (sometimes several times if I tightened it a hair to far over), cursing and swearing and fussing with those stupid plastic camshaft holders that did the trick. The last time I did it I got the timing bang on the first try with no swearing! Complete fluke of course... :) That was the timing post quick-and-and-dirty ring job and I was very motivated to bolt the car back together and break in the new rings... ;)

Drove the car down the highway this evening to test out the O2 and Knock sensor - seems better. Ignition leads are next on the bucket list - clutch slave rebuild kit plus slave speedbleeder mod are up next this week.
 
Back