Is you Knock sensor taking performance down for the count?

[LGAFF]

Interesting to note that after running 12.4s @ 117.5mph and then adding a lightweight flywheel I saw a decrease in performance. I had been running 12.6s at 114.5-115 range.....I then noticed my knock sensor was unplugged, problem solved right? wrong! after plugging it in I saw no change, then last year I purchased a Tech 1.....my car was pulling 10 degrees of timing,constantly and in fact on hard acceleration I was pulling even more.....

So I went down to my local parts store and purchased a ZR-1 specific replacement...Pete had warned me to check it by OHMing the sensor and that a ZR-1 sensor should OHM at 3.0-4.0....wrong resistence; this sensor from O"Reillys tested at 99.

Then I tested the one on the spare motor, bad too..no reading

Now keep in mind my 92 has 23K miles on it....and even with the timing retard it seems very seat of the pants fast.....so how many of you weekend warriors at the strip are losing .2-.5 from timing retard.

Time to pull the Fluke meter out and check yours....oh, by the way GM discontinued the Z knock sensor....they are still availble from some vintage parts places, Jerrys, and I imagine Kurts too.

Thanks to Pete for the heads up!

[mike100]

I have not observed much pulled timing on my tech1 data. maybe a degree or two and a small ping when it is hot, but mostly it is showing the full amount of timing when I have done the recording under accel at the data rate the scanner is capable of.

The biggest question is why does it pull timing when you disconnect the sensor? Does it have to have a grounded signal to run properly?

I can't recall, does it list a timing retard number? I do know that the total timing shown is 6 degrees less because the raw value does not include the 6 degrees of lead cast into the sensor mount/crank notch.

I will check it next time I change the oil out of curiosity.

[LGAFF]

I do know that are cars with known false knock issues that use a resistor to create a false signal that the sensor is present to thus avoiding "False Knock" and avoiding timing pull from the sensor disconnect.

[LGAFF]

Early 85=89 TPI uses a 3.9K knock sensor, wonder if that would work for us?

[Pete]

If you unplug the sensor the ECM thinks the sensor is bad basicly there's no singal/voltage going to the ECM and will retard timing so you don't blow up/hurt motor.

Pete

[Paul Workman]

Tuning the knock sensor...

This may open the door to using currently available knock sensors... What say you Tuner Pro gurus?

C-link

P.

[XfireZ51]

Paul,

We'd need to determine if the EST module in the ZR is specific to the motor(which I suspect due to the completely different motor architecture)
or it works w another KS. Perhaps Marc H. or Jerry could shed some light on that. The knock strategy in the stock calibration is pretty aggressive and clearly designed for better safe than sorry. Knock parameters deal w amount of retard,
Duration and decay rates. These can be reduced typically. If you've data logged a dyno tuning session, you'll see that the LT-5 isn't prone to knock at the top end. Knock shows up right around torque peak, between 4-5500 rpm. In my last tuning session I had 34d for top end advance w no knock. However, power was down. So no sense in keeping it there.
Proper installation of the KS is important(ie overtightening) to avoid false knock.

[XfireZ51]

"Now keep in mind my 92 has 23K miles on it....and even with the timing retard it seems very seat of the pants fast.....so how many of you weekend warriors at the strip are losing .2-.5 from timing retard."

That's why I datalog my runs.

[LGAFF]

Well the knock sensor in my 90 motor was bad too, the one from 966......maybe that explains the drop in performance on the dyno.

So knock sensor in my 92 dead at 23K miles, in my 90 and 143K and the spare motor at 40K

[gbmidyear66]

I'm confused... I had understood (incorrectly?) that the LT5 does a self test of the knock sensor / circuitry at start up. If the seonsor is inoperative - presumably it will indicate a DTC?