volkl77

Ok... here we go again...fingers X'd this time

So far I've tested both new and old sensors and the values were very similar. Gut feeling told me that something wasn't quite right. Here's what happened.

1.Resistance
Tested with leads attached and left for a minimum of 1 minute to stabilize.

New 550 k Ohm
Old 549 k Ohm

No significant difference here.

2. Continuity
Neither sensor showed continuity across both terminals. This was contrary to what I've seen here on the forums so far.

3. AC Voltage
This is where things got interesting. In hand there seemed to be little difference the values between new and old when tapped to simulate engine knock. I figured a more controlled test was required. I rigged up a plate of steel with two threaded holes in it. One to mount the KS with its own bolt,torqued to 15 ft/lbs, and the rod and slide section of my slide hammer in the second hole. With the meter connected and set to hold peak values, I dropped the hammer from 6".

Old sensor: peaks of around 65 to 70 millivolts
New sensor: peaks of around 35 millivolts

I'm going to assume that the AC voltage generated by the sensor when it detects engine knock is the signal that the ECU uses. I would also assume that the ECU is looking for a value within a certain range to determine if the sensor is functioning. The question now is what range of voltage values is the ECU looking for? Is the old sensor "worn" and as a result, too sensitive?

I installed the new sensor and reset the ECU. Started and ran the engine till it warmed up, rev'd a few times (I could just be hoping but it seemed to spool up faster and smoother), shut down then checked the ECU again. No 0304...just 0505. So far so good. We'll see what happens out on the road when these thunderstorms pass.

More to come

Update

Its been 2 days, and after resetting the ECU, the engine seems like its back to normal. Mileage is improving and there is a definitite increase in power and throttle response.
I suppose the bottom line here is even if the sensor seems to test ok, the ECU knows better. Dang AI!

Nismo87SE

To my knowledge no one has ever done a test of this nature. While the resistance tested out ok, the response wasn't as precise as the new model. This is the samething that happens to o2 sensors, as they get old they cycle between rich/lean slower.