My knock sensor wire was barely holding on to the wire connector where it plugs into the wiring harness in front of the rats nest, it broke off when doing my engine rebuild. i cut the wire and spliced it back together with male and female wire connectors, no more factory wire connector. well now the knock sensor has not worked since pulling out the blown up engine. for the most part there is no knock sensor reading on the pfc commander, sometimes it will show a max reading of 1, 3 or 5 nothing higher than 5. it worked good before the wire broke off with the old motor. please help. thanks!

 

Tom,

How do you tell if the knock sensor is working? I've used an ohmmeter and also tapped into the wire with a voltmeter set on both DC and AC volts, and hit the block (easily) with a metal hammer, but neither my old nor my brand-new knock sensor has ever reacted at all. I think the element is a piezoelectric crystal that generates a voltage like a microphone when it "hears" something, but how the heck can you read it? With an oscilloscope? What do the PFC numbers mean? (I have a stock PCME.)

 

i dont know. all i know is with my old motor it would read higher numbers on the commander, like 30 or 40 etc. now it doesnt really read anything. could the rebuilt motor really produce no knock, I doubt it. could it have something to do with mounting of the knock sensor on the new engine or the wire fix job somehow not good enough, I dont know.

 

What does the Commander read if you tap on the engine supports with a metal object as described in the factory manual? On the stock system (stock PCME), the manual says to test it as follows:

1. Connect the negative terminal of a voltmeter to the "MEN" terminal in the data link connector. Connect the positive terminal to a chassis ground.
2. Connect a jumper wire between the "TEN" and "GND" terminals in the data link connector.
3. Turn the ignition switch to "ON."
4. Lightly tap the engine hanger with a (metal) hammer.
5. Verify that the voltmeter indicator moves.
6. Turn ignition switch off.

Not sure I did it that way... maybe I'll try it again. Might have had the voltmeter terminals reversed - duh! Don't know how you do it with a PFC.

 

I believe that is a shielded wire and requires special care when splicing it, otherwise you will short the inner conductor and render the sensor inoperable. I would suggest checking your splice.

 

^Yep, there's a shield around the wire for most of the run. If the shield is crimped into the core you'll get nothing.

The shield is supposed to go to ground to help reduce noise in the signal since it's very low voltage.

Best way to test is just reading the knock with the commander. The TEN/Ground stuff doesn't work with the PFC.

Dale

 

i ran into this issue off a car that came from the dealership after they did an engine R+R, found that they crimped the connector over the outer shield core wire so the signal was constantly grounded and the ECU was tossing a code for the knock sensor since it wasn't seeing an open, versus a grounded signal. the knock sensor, O2 and crank triggers all have shielded wires.

 

i dont know I have not tried any tests. Thanks for that info. the commander just displays a number while its running, the higher the number the more knock.

 

when I did the wire fix i peeled back the shielding so when i connected the wire there was no sheildini around that part of the wire. and if i remember correctly, there is no shielding around the wire harness side of the factory connector, that wire had no shielding. the shielding ended at the connector. so i dont see how it would get grounded. is that right or wrong? thanks guys.

 

the shielding starts on the engine harness side of the connector and ends about 1" from the connector, leaving a little bit of room but not much. there is no shielding on the sensor side of the connector iirc.

 

what about on the O2 sensor, it the sheilding only on the engine side of the harness, b/c i was under the impression it was sheilded on the sensor side also (for the O2...)

 

I think the knock sensor is actually shielded on both sides of the connector.

After looking at my wire again, I see it was the connector on the engine harness side that actually fell off the wire, so I attached a female wire connector to it and just plugged it back into the connector coming from the knock sensor. The connector I used is a little gold plated connector, same as this one pictured here http://shop.advanceautoparts.com/wcs...1_pri_larg.jpg. Could this not be a good connector to use? It does plug together tightly.

My knock reading on the PFC commander shows a 4 right after starting the car, but that's the only time it will show a reading. I tried hitting the engine hook with a hammer a bunch of times and nothing showed up for knock, also tapped the big knock sensor bolt, nothing.

Any thoughts on the wire connection? thanks.

 

ohm test both sides of the connector to ground. if there is no continuity or any reading under a few k ohms then try a different sensor.

 

Does a rotary engine "knock" or "ping" sound like that in a piston engine? I've never heard that sound from our '94 FD.

 

Could you help me out a little more with the ohm test for the sensor? Sorry I've never done a ohm test before on wires, not sure what to do. I do have a volt meter which I think does ohms too.

The knock sensor wire coming from the engine harness looks like a thick brown wire, but when you open it up there is a wire that surrounds a little white wire. I had my wire connector connected to this white wire, and is the other wire in there the shielding?

I found a spot where the white wire was maybe exposed to the other shielded wire so I assume this could have shorted something out? Would it mess up the PFC or the sensor if these two wires touched each other?

I used new wire connectors and cut back the brown wire covering and pulled back the shielded wire and made sure the two weren't touching each other as far as I could see but the knock sensor still doesn't seem to work, it just shows a 1 or a 3 or a 4 when the car starts. nothing shows up when hitting the engine hook with a hammer. I also cut off the factory wire connector on the sensor side to rule that connector out as a possible problem. thanks again.

 

set your dial indicator to the ohm figure and hook it up like you would as a voltmeter. hook up the negative lead to a good solid ground on the engine block, hook up the red lead to each side of the connector on the harnesses and get a reading from the ohm meter. if the signal is infinite(zero resistance or an a complete open circuit) then the meter reading will not change from it's original reading. minor resistance will usually read in meg-ohms which will have a "M" after the reading, kilo ohms(or 1000 ohms) will have a "K" after it and hundred ohms will be listed as a raw figure on the meter. i'll have to check and see what the typical resistance reading is of the knock sensor but there shouldn't be any real measurable resistance in the knock sensor circuit to ground on the engine harness side

if you have a basic meter it will have several settings, like 10, 100, 1000, etc, start with 10 and get a reading and work your way up. multiply the reading times the setting on the meter for your total reading figure.

so, in short, you should have a reading of more than about 5,000 ohms, or 5k ohms for short. usually wiring harnesses will get some noise and give a reading in the meg-ohms when you hook them up, this is fairly normal.

yes, the inside wire is the one you need to tap into not the outer frayed type wire surrounding it.

yes, even a single strand of wire that gets into your connection from the outer shield wire will drastically effect the signal. say if the wires got pinched under the alternator and smashed then it would kill the wire signal.


i would just do the ohm test and see what figure you get, if it looks ok then i would just replace the sensor because they are easy to find and fail regularly(however usually they don't flat out die but they will give higher than normal knock readings versus low as the gel coating liquifies and leaks out of the sensor).

 

Thanks alot for all that info. I will do the test and probably have to find a used knock sensor to try and get it working again.

 

So somehow my knock sensor just started working properly it seems. I can't really explain why. I just got done installing a crane hi-6dsr and wired up a relay to activate the built in 2 step rev limiter (which works great), and now the knock sensor seems like it works like it should. I didn't do anything to the knock sensor or its wire, so its very strange, but glad it works now.