I recently just replaced my optispark. Car has 80k+ miles. I can't win. Now one or both of the knock sensors are DOA. With the ECM connector removed and SHORTLY AFTER running the car for just a few seconds, the circuit will read over 6k ohms. Not good. Acceptable value to the ECM pin is supposed to be 3300 to 4500 ohms. If the car has sat for a few hours, the pin will read 2500 ohms, still not good!
So, here's the deal..

I propose to disconnect the wire at the ECM and, with the proper resistors, permanently tie off the sensor input to be within the "acceptable range".

Yeah, yeah, I know, the sensors are there for a reason. I feel like being a rebel, but was looking for some more input before I modify the harness and heatshrink all the goodies. What's a little knock? I can't ever recall having a bad tank of gas, and the rest of systems and motor are in excellent working order.

 

I think it depends on how "big" a "little" knock gets. Before you realize it and throttle down to relieve cylinder pressure.

You paid those 6 figure engineers to discover how to protect your engine from the harm a "little" knock can cause.
This 'may' be a case when a little knowledge (no insult meant) is more dangerous than a lot.
You obviously know how the system works so you're smart enough to not defeat it.
A knock sensor doesn't cost that much.

Experience is the best teacher but she sends horrific bills.

 

Just run 104 octane race gas in it all the time and you won't have to worry about knock....

 

The knock sensor's output is not resistance based. It's DC resistance (what you're measuring with your meter) can easily vary within the ranges you've stated.

To the ECM, the knock sensor signal appears as a small AC voltage source. To truely determine if the sensor is functioning properly, you need to observe the sensor's output with an oscilloscope and /or spectrum analyzer. With the engine running, tap the exhaust header with a metal object ( I used a 3/4" x 12" ratchet extension). You'll typically see the sensor emit frequencies between 4KHz and 8KHz.

Another method is to use a ECM scan tool. Observe the knock counts with the engine running, tap the exhaust header and knock counts should increment if the system is functioning properly.

 

Did it set a code? How much knock retard are you seeing when you scan it? I tend to agree with the above thoughts-make it work correctly. It may just save a few $$$ in the long run!
If you are getting obviously false knock retard then try to rule out the source; anything loose/banging around under acceleration. Another suggestion I've read before to rule out false knock-fill it up with gas at the airport-if the knock goes away then it wasn't false......

 

As much of a PITA that it is to find the source of the problem it would be worth your while to do so. Over the years I have had a few run ins with knock sensors. There is one difficult question that always needs to be answered: "is this real knock/detonation, or is it imaginary?" If it goes away with the use of 100 octane unleaded (the 104 unleaded is hard on cats), then it is real and tuning may eliminate the problem. Beware that in the case of the knock sensor going off even with high test fuel in the car that does NOT mean the issue is imaginary.

Here is an example of imaginary knock that I had recently: my knock sensors were going off even at idle and just cruising down the road. Given everything else, I knew that imaginary it had to be - I even went drag racing with it like that, although it did cost me some performance. Later I did discover the problem - it was an idler pulley with a bad bearing. On a whim I removed the serpentine belt to see if the problem went away, it did so I started checking out all the accessory drive items until I found the bad pulley/bearing. Another cause in the past has been broken studs on the exhaust manifold collector. When I had a severly carbon tracked opti cap and rotor the spark knock system went nuts.

Here is an example of tuning related knock retard: yesterday during my chassis dyno tuning session there was six degrees of ignition advance being pulled out of the timing upon initial WOT at low rpm. When I changed my low rpm WOT advance from 34 degrees to 31 degrees the problem went away.

Good luck, and whatever you do don't foolishly hurt your engine by ignoring the knock sensors call to attention.

Thomas

 

I am not condoning taking the KS off, but I just had to too.

After painfully looking into every detail I could, and not being able to defeat the KS I backed my timing all the way back to 25.3* and bam the knock went away.

I went to the track and could only get a best of 105 MPH and 13.2 sec 1/4. Upon the request of some people here and at the track I disconnected the knock sensor. With a chip used for our setups I am running 36.2* of advance with no problems and gobs more power than before.

My case may be special, and I am not condoning or encouraging the removal.




[Modified by ski_dwn_it, 9:18 PM 5/31/2002]

 

http://forums.corvetteforum.com/zerothread?id=289155
:cheers:


[Modified by vette_tweak, 9:00 PM 6/1/2002]