I understand their function, but what makes them work? Can I do away with them? How? What will I loose if I do? Any help extremely appreciated.

They are microphones filtered to the specific knock frequency of the engine. You need them more than you need sex. Without them you risk (and a high probablity) of detonation and engine failure.

[This message has been edited by TK (edited 05-26-2003).]

TK. Thanks for the reply!
Like I said I understand their function. They provide a signal proportional in frquency and amplitude of a knock (useally ingnition predetination) which causes the timing to retard (and possibly the fuel mixture to become richer). This allows the engine to run as effeciently as possible without knocks. I guess what I really want to know is if I disconnect them does the PCM know and react to that condition? If the "microphone" element is ceramic it wouldn't draw any current and the PCM couldn't know it was missing. If its carbon the PCM could know. Somewhere in a manual I read that the engine wouldn't run without them but I beleive this to be toro caca. If I just disconnect them will the engine still function at full power but possibly knock like hell?

It'll run just fine without the sensor connected. Buddy of mine accidentally left one disconnected, broke a piston and all that fun. Me? I think it's cheaper to leave it functional.

JazzMan

Jazzman, Thanks thats what i needed! Question about that broken piston. He left just one disconnected? Does the PCM actually adjust timing and mixture on a per cylinder or bank basis. Why do you need 2 sensors? I understand that 2 is always better than one because it's more likely to pick up sound, but wouldn't all the cylinders knock if the knock was due to advanced timing or lean mixture? Just like to know, kinda anal I guess!

Not necessarily. Take for instance an injector leans out or doesn't fire altogether. That's why you would have 1 in each bank.

------------------
As Alice Cooper once said...

quote Feed My.... Frankenstein!

85frankenstien, follow you and agree about the leaning out. Faster combustion equals knock. that's probably why they use one for each bank. But..no combustion equals no knock at all just shitty 5 cylinder performance.

Touche! Ya got me on that point, but knowing boosted engines as well as I do, there's bunch of things that could go wrong. I'd feel a helluva lot safer knowing that the ECU is monitoring both banks. I'd hate to smoke a piston and have to rip the motor down for a ~$30 - $40 part.

85frankenstien, I'm with ya there! Thanks for the help!

If it is disconnected it will throw a code and subtract a fixed amount of advance (or add in a fixed amount of retard....). I think it's about 15 degrees. I'd have to look. Since wot is around 10-12 degrees you can see the result. It will loose power. What's bad is faking it out.

I am surprised that one guy sucked a piston since there is no real advance with it disconnected although maybe he faked it out or disabled the error code.

The PCM knows it's gone because there is a shut resistor in the sensor that is the bottom leg of a voltage divider. Since it's not there, the DC bias floats up and the PCM senses that and throws the code. Same if it's grounded, the bias is too low and it throws the code. That's why you can't use just any KS.

With the code it won't even do the octane test.

If I get a chance tonight I'll see what the retard is on a KS fault in the S2 code.

[This message has been edited by TK (edited 05-26-2003).]

actually Terry, I believe Kevin told me that it takes out 5 degrees of timing across the board. So in other words about 20 hp. I drove my car for almost a year with my knock sensor hooked up incorrectly, and that along with my intake air temp sensor messed up, and my coolant temp sensor not hooked up right, it was pulling 9 degrees of timing. Needless to say, it never knocked, but I was loosing almost 40 hp at the wheels. Once all this stuff was fixed I went from a 12.004 to 11.633 in the 1/4.

So...on an S2...

Max retard without a malf is 15 degrees.

With a malf, it's 2-2.5 degrees but it looks like it goes into Piston Protection mode which reduces the advance so it has the same affect.

Ah, life was so much easier on the S1's......

There is only one pin on the sensor, therefore the return has to be the chassic. I cannot measure any continuity between this pin and chassic ie the resistance is infinite. If there is shunt resistance (assuming the word "shut" is a typo in your reply) why can't I measure it? A carbon microphone would act like a resistor, ceramic, I guess would be more like capapcitive. If the "microphone is ceramic and part of an oscillator removing it could be detected by the PCM because the frequency of the oscillation would change or the oscillator would completely stop. My manual (Haynes) says the sensor can be checked by removing it from the block, leaving it attatched to voltage, and measuring the AC voltage with a AC voltmeter from the pin to ground while submitting it to vibration. I'm thinking an AC millivoltmeter or an oscilloscope might be more appropriate, since I can't imagine a very significant voltage fluctuation. I don't yet have DC power connected so I can't take voltage readings to find out if the voltage rises when disconnecting the sensor, or if there is an oscillation without vibration. Your statement about the set retarted timing makes sense, this is the type of thing I would expect to see, but I'm still puzzled on how the hell they do it. Please let me know what your research reveals and thanks again for your help!

quote Originally posted by fieroX: actually Terry, I believe Kevin told me that it takes out 5 degrees of timing across the board. So in other words about 20 hp. I drove my car for almost a year with my knock sensor hooked up incorrectly, and that along with my intake air temp sensor messed up, and my coolant temp sensor not hooked up right, it was pulling 9 degrees of timing. Needless to say, it never knocked, but I was loosing almost 40 hp at the wheels. Once all this stuff was fixed I went from a 12.004 to 11.633 in the 1/4.

What calibration did Kevin use? I just looked at the 98.

Thanks TK. I'll get my DVM out tomorrow evening and check again. Sound like I may have a bad sensor! I'll compare bank 1 to bank 2 ect. I know I can fix this with common sense and less analization, but I can't stand not knowing how it works! Thanks for you patience!

Patient? Me? Lord, tell my wife that!

Both should read the same. Let us know what you find.

100KOhm's exactly! Don't know what the hell I did the first time! And I discovered that if you have an analog meter on your DMM you can actually see changes in resistance measurements when adding vibration (digital readout doesn't respond to instantaneous fluctuations).

Cool!

OK, TK, Now please allow grasshopper to move on to next lesson. If 100K resistors are installed to complete the voltage divider to fool the PCM will the PCM then advance the timing until bad stuff happens or will the PCM merely be unable to retard the timing IF something bad occurs (lean mixture, shitty gas, to much boost, EGR falls off and is misteriously replaced with plugs)? Thanks in advance!

Right, if you put 100k resistors in there you will not detect knock or pull timing when the bad juju happens.

Those sensors really bug you don't they......you need to calm down. You're scared and confused......

Removing the knock sensors is like defeating the safety on a hair trigger.

[This message has been edited by TK (edited 05-27-2003).]

Safety? Never heard of it. My hog leg has a hair trigger and there's nothing safe about it! Calm is for tokers! No, I'm OK. Where the Hell's my axe. I like them fine. I'm going to put a couple on my door, yeah, a PLC connected to my PC, a solonoid, a string, my 12 double cut off, yeah that's the ticket! Get away I tell you I'm OK....